Model save
Browse files- README.md +71 -0
- all_results.json +7 -0
- config.json +30 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +213 -0
- train_results.json +7 -0
- trainer_state.json +2417 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: DatPySci/pythia-1b-sft-full
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: pythia-1b-kto-iter0
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# pythia-1b-kto-iter0
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [DatPySci/pythia-1b-sft-full](https://huggingface.co/DatPySci/pythia-1b-sft-full) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.2591
|
| 18 |
+
- Rewards/real: 0.0604
|
| 19 |
+
- Rewards/generated: -1.0267
|
| 20 |
+
- Rewards/accuracies: 0.9460
|
| 21 |
+
- Rewards/margins: 1.0871
|
| 22 |
+
- Logps/generated: -570.8114
|
| 23 |
+
- Logps/real: -468.1696
|
| 24 |
+
- Logits/generated: 0.2253
|
| 25 |
+
- Logits/real: -0.2820
|
| 26 |
+
|
| 27 |
+
## Model description
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Intended uses & limitations
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training and evaluation data
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training procedure
|
| 40 |
+
|
| 41 |
+
### Training hyperparameters
|
| 42 |
+
|
| 43 |
+
The following hyperparameters were used during training:
|
| 44 |
+
- learning_rate: 5e-07
|
| 45 |
+
- train_batch_size: 4
|
| 46 |
+
- eval_batch_size: 4
|
| 47 |
+
- seed: 42
|
| 48 |
+
- distributed_type: multi-GPU
|
| 49 |
+
- num_devices: 2
|
| 50 |
+
- gradient_accumulation_steps: 8
|
| 51 |
+
- total_train_batch_size: 64
|
| 52 |
+
- total_eval_batch_size: 8
|
| 53 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
+
- lr_scheduler_type: linear
|
| 55 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 56 |
+
- num_epochs: 1
|
| 57 |
+
|
| 58 |
+
### Training results
|
| 59 |
+
|
| 60 |
+
| Training Loss | Epoch | Step | Validation Loss | Rewards/real | Rewards/generated | Rewards/accuracies | Rewards/margins | Logps/generated | Logps/real | Logits/generated | Logits/real |
|
| 61 |
+
|:-------------:|:-----:|:----:|:---------------:|:------------:|:-----------------:|:------------------:|:---------------:|:---------------:|:----------:|:----------------:|:-----------:|
|
| 62 |
+
| 0.2932 | 0.38 | 300 | 0.2962 | 0.0718 | -0.7855 | 0.9220 | 0.8572 | -568.3989 | -468.0556 | 0.2554 | -0.2530 |
|
| 63 |
+
| 0.2689 | 0.77 | 600 | 0.2591 | 0.0604 | -1.0267 | 0.9460 | 1.0871 | -570.8114 | -468.1696 | 0.2253 | -0.2820 |
|
| 64 |
+
|
| 65 |
+
|
| 66 |
+
### Framework versions
|
| 67 |
+
|
| 68 |
+
- Transformers 4.38.1
|
| 69 |
+
- Pytorch 2.2.1
|
| 70 |
+
- Datasets 2.17.1
|
| 71 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.31094988259020595,
|
| 4 |
+
"train_runtime": 9866.2092,
|
| 5 |
+
"train_samples_per_second": 5.068,
|
| 6 |
+
"train_steps_per_second": 0.079
|
| 7 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DatPySci/pythia-1b-sft-full",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": true,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 2048,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 8192,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 8,
|
| 20 |
+
"num_hidden_layers": 16,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rotary_emb_base": 10000,
|
| 23 |
+
"rotary_pct": 0.25,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "bfloat16",
|
| 26 |
+
"transformers_version": "4.38.1",
|
| 27 |
+
"use_cache": false,
|
| 28 |
+
"use_parallel_residual": true,
|
| 29 |
+
"vocab_size": 50304
|
| 30 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.38.1",
|
| 6 |
+
"use_cache": false
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a7a9eee68c707be5a4b8e4666aaaa69c988bfdd5aaa1db44ffcc7b18f046603
|
| 3 |
+
size 2023586384
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,213 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<|padding|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"50254": {
|
| 21 |
+
"content": " ",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": false
|
| 27 |
+
},
|
| 28 |
+
"50255": {
|
| 29 |
+
"content": " ",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": false
|
| 35 |
+
},
|
| 36 |
+
"50256": {
|
| 37 |
+
"content": " ",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": false
|
| 43 |
+
},
|
| 44 |
+
"50257": {
|
| 45 |
+
"content": " ",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": false
|
| 51 |
+
},
|
| 52 |
+
"50258": {
|
| 53 |
+
"content": " ",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": true,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": false
|
| 59 |
+
},
|
| 60 |
+
"50259": {
|
| 61 |
+
"content": " ",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": true,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": false
|
| 67 |
+
},
|
| 68 |
+
"50260": {
|
| 69 |
+
"content": " ",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": true,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": false
|
| 75 |
+
},
|
| 76 |
+
"50261": {
|
| 77 |
+
"content": " ",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": true,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": false
|
| 83 |
+
},
|
| 84 |
+
"50262": {
|
| 85 |
+
"content": " ",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": true,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": false
|
| 91 |
+
},
|
| 92 |
+
"50263": {
|
| 93 |
+
"content": " ",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": true,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": false
|
| 99 |
+
},
|
| 100 |
+
"50264": {
|
| 101 |
+
"content": " ",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": true,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": false
|
| 107 |
+
},
|
| 108 |
+
"50265": {
|
| 109 |
+
"content": " ",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": true,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": false
|
| 115 |
+
},
|
| 116 |
+
"50266": {
|
| 117 |
+
"content": " ",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": true,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": false
|
| 123 |
+
},
|
| 124 |
+
"50267": {
|
| 125 |
+
"content": " ",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": true,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": false
|
| 131 |
+
},
|
| 132 |
+
"50268": {
|
| 133 |
+
"content": " ",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": true,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": false
|
| 139 |
+
},
|
| 140 |
+
"50269": {
|
| 141 |
+
"content": " ",
|
| 142 |
+
"lstrip": false,
|
| 143 |
+
"normalized": true,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": false
|
| 147 |
+
},
|
| 148 |
+
"50270": {
|
| 149 |
+
"content": " ",
|
| 150 |
+
"lstrip": false,
|
| 151 |
+
"normalized": true,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": false
|
| 155 |
+
},
|
| 156 |
+
"50271": {
|
| 157 |
+
"content": " ",
|
| 158 |
+
"lstrip": false,
|
| 159 |
+
"normalized": true,
|
| 160 |
+
"rstrip": false,
|
| 161 |
+
"single_word": false,
|
| 162 |
+
"special": false
|
| 163 |
+
},
|
| 164 |
+
"50272": {
|
| 165 |
+
"content": " ",
|
| 166 |
+
"lstrip": false,
|
| 167 |
+
"normalized": true,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false,
|
| 170 |
+
"special": false
|
| 171 |
+
},
|
| 172 |
+
"50273": {
|
| 173 |
+
"content": " ",
|
| 174 |
+
"lstrip": false,
|
| 175 |
+
"normalized": true,
|
| 176 |
+
"rstrip": false,
|
| 177 |
+
"single_word": false,
|
| 178 |
+
"special": false
|
| 179 |
+
},
|
| 180 |
+
"50274": {
|
| 181 |
+
"content": " ",
|
| 182 |
+
"lstrip": false,
|
| 183 |
+
"normalized": true,
|
| 184 |
+
"rstrip": false,
|
| 185 |
+
"single_word": false,
|
| 186 |
+
"special": false
|
| 187 |
+
},
|
| 188 |
+
"50275": {
|
| 189 |
+
"content": " ",
|
| 190 |
+
"lstrip": false,
|
| 191 |
+
"normalized": true,
|
| 192 |
+
"rstrip": false,
|
| 193 |
+
"single_word": false,
|
| 194 |
+
"special": false
|
| 195 |
+
},
|
| 196 |
+
"50276": {
|
| 197 |
+
"content": " ",
|
| 198 |
+
"lstrip": false,
|
| 199 |
+
"normalized": true,
|
| 200 |
+
"rstrip": false,
|
| 201 |
+
"single_word": false,
|
| 202 |
+
"special": false
|
| 203 |
+
}
|
| 204 |
+
},
|
| 205 |
+
"bos_token": "<|endoftext|>",
|
| 206 |
+
"chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
|
| 207 |
+
"clean_up_tokenization_spaces": true,
|
| 208 |
+
"eos_token": "<|endoftext|>",
|
| 209 |
+
"model_max_length": 2048,
|
| 210 |
+
"pad_token": "<|endoftext|>",
|
| 211 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 212 |
+
"unk_token": "<|endoftext|>"
|
| 213 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.31094988259020595,
|
| 4 |
+
"train_runtime": 9866.2092,
|
| 5 |
+
"train_samples_per_second": 5.068,
|
| 6 |
+
"train_steps_per_second": 0.079
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2417 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.99968,
|
| 5 |
+
"eval_steps": 300,
|
| 6 |
+
"global_step": 781,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"grad_norm": 38.75,
|
| 14 |
+
"learning_rate": 6.329113924050633e-09,
|
| 15 |
+
"logits/generated": 0.1862616091966629,
|
| 16 |
+
"logits/real": -0.24196800589561462,
|
| 17 |
+
"logps/generated": -538.626953125,
|
| 18 |
+
"logps/real": -411.1020202636719,
|
| 19 |
+
"loss": 0.5,
|
| 20 |
+
"rewards/accuracies": 0.0,
|
| 21 |
+
"rewards/generated": 0.0,
|
| 22 |
+
"rewards/margins": 0.0,
|
| 23 |
+
"rewards/real": 0.0,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01,
|
| 28 |
+
"grad_norm": 32.0,
|
| 29 |
+
"learning_rate": 3.1645569620253166e-08,
|
| 30 |
+
"logits/generated": 0.34831082820892334,
|
| 31 |
+
"logits/real": -0.12466613948345184,
|
| 32 |
+
"logps/generated": -534.7514038085938,
|
| 33 |
+
"logps/real": -401.6441345214844,
|
| 34 |
+
"loss": 0.4933,
|
| 35 |
+
"rewards/accuracies": 0.390625,
|
| 36 |
+
"rewards/generated": 0.028874225914478302,
|
| 37 |
+
"rewards/margins": -0.008529577404260635,
|
| 38 |
+
"rewards/real": 0.020344652235507965,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.01,
|
| 43 |
+
"grad_norm": 29.125,
|
| 44 |
+
"learning_rate": 6.329113924050633e-08,
|
| 45 |
+
"logits/generated": 0.37132272124290466,
|
| 46 |
+
"logits/real": -0.12087633460760117,
|
| 47 |
+
"logps/generated": -555.05859375,
|
| 48 |
+
"logps/real": -418.5079040527344,
|
| 49 |
+
"loss": 0.4941,
|
| 50 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 51 |
+
"rewards/generated": 0.00217894883826375,
|
| 52 |
+
"rewards/margins": -0.003474087920039892,
|
| 53 |
+
"rewards/real": -0.0012951357057318091,
|
| 54 |
+
"step": 10
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.02,
|
| 58 |
+
"grad_norm": 31.375,
|
| 59 |
+
"learning_rate": 9.49367088607595e-08,
|
| 60 |
+
"logits/generated": 0.3862614035606384,
|
| 61 |
+
"logits/real": -0.1023840680718422,
|
| 62 |
+
"logps/generated": -564.0845947265625,
|
| 63 |
+
"logps/real": -435.2381286621094,
|
| 64 |
+
"loss": 0.4815,
|
| 65 |
+
"rewards/accuracies": 0.581250011920929,
|
| 66 |
+
"rewards/generated": -0.025657888501882553,
|
| 67 |
+
"rewards/margins": 0.04708629846572876,
|
| 68 |
+
"rewards/real": 0.021428411826491356,
|
| 69 |
+
"step": 15
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.03,
|
| 73 |
+
"grad_norm": 30.75,
|
| 74 |
+
"learning_rate": 1.2658227848101266e-07,
|
| 75 |
+
"logits/generated": 0.3665524125099182,
|
| 76 |
+
"logits/real": -0.06457304954528809,
|
| 77 |
+
"logps/generated": -558.3909301757812,
|
| 78 |
+
"logps/real": -439.05072021484375,
|
| 79 |
+
"loss": 0.5034,
|
| 80 |
+
"rewards/accuracies": 0.512499988079071,
|
| 81 |
+
"rewards/generated": 0.007058300077915192,
|
| 82 |
+
"rewards/margins": -0.009866954758763313,
|
| 83 |
+
"rewards/real": -0.0028086546808481216,
|
| 84 |
+
"step": 20
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.03,
|
| 88 |
+
"grad_norm": 31.5,
|
| 89 |
+
"learning_rate": 1.5822784810126582e-07,
|
| 90 |
+
"logits/generated": 0.39779749512672424,
|
| 91 |
+
"logits/real": -0.1216607466340065,
|
| 92 |
+
"logps/generated": -565.197509765625,
|
| 93 |
+
"logps/real": -439.4720153808594,
|
| 94 |
+
"loss": 0.4893,
|
| 95 |
+
"rewards/accuracies": 0.5,
|
| 96 |
+
"rewards/generated": 0.0025315589737147093,
|
| 97 |
+
"rewards/margins": -0.010492642410099506,
|
| 98 |
+
"rewards/real": -0.007961079478263855,
|
| 99 |
+
"step": 25
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.04,
|
| 103 |
+
"grad_norm": 32.5,
|
| 104 |
+
"learning_rate": 1.89873417721519e-07,
|
| 105 |
+
"logits/generated": 0.3631184995174408,
|
| 106 |
+
"logits/real": -0.15178892016410828,
|
| 107 |
+
"logps/generated": -563.345947265625,
|
| 108 |
+
"logps/real": -444.08758544921875,
|
| 109 |
+
"loss": 0.4822,
|
| 110 |
+
"rewards/accuracies": 0.53125,
|
| 111 |
+
"rewards/generated": -0.02776198461651802,
|
| 112 |
+
"rewards/margins": 0.05192292481660843,
|
| 113 |
+
"rewards/real": 0.024160942062735558,
|
| 114 |
+
"step": 30
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04,
|
| 118 |
+
"grad_norm": 32.5,
|
| 119 |
+
"learning_rate": 2.2151898734177212e-07,
|
| 120 |
+
"logits/generated": 0.353170782327652,
|
| 121 |
+
"logits/real": -0.08487684279680252,
|
| 122 |
+
"logps/generated": -567.288330078125,
|
| 123 |
+
"logps/real": -426.352783203125,
|
| 124 |
+
"loss": 0.478,
|
| 125 |
+
"rewards/accuracies": 0.581250011920929,
|
| 126 |
+
"rewards/generated": -0.05789921432733536,
|
| 127 |
+
"rewards/margins": 0.08254577219486237,
|
| 128 |
+
"rewards/real": 0.02464655414223671,
|
| 129 |
+
"step": 35
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.05,
|
| 133 |
+
"grad_norm": 28.0,
|
| 134 |
+
"learning_rate": 2.5316455696202533e-07,
|
| 135 |
+
"logits/generated": 0.3951062560081482,
|
| 136 |
+
"logits/real": -0.10047361999750137,
|
| 137 |
+
"logps/generated": -584.459228515625,
|
| 138 |
+
"logps/real": -481.53759765625,
|
| 139 |
+
"loss": 0.4788,
|
| 140 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 141 |
+
"rewards/generated": -0.01975923776626587,
|
| 142 |
+
"rewards/margins": 0.056668005883693695,
|
| 143 |
+
"rewards/real": 0.036908768117427826,
|
| 144 |
+
"step": 40
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.06,
|
| 148 |
+
"grad_norm": 29.375,
|
| 149 |
+
"learning_rate": 2.848101265822785e-07,
|
| 150 |
+
"logits/generated": 0.38055142760276794,
|
| 151 |
+
"logits/real": -0.10685946047306061,
|
| 152 |
+
"logps/generated": -590.1273193359375,
|
| 153 |
+
"logps/real": -475.4049377441406,
|
| 154 |
+
"loss": 0.4803,
|
| 155 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 156 |
+
"rewards/generated": -0.034241292625665665,
|
| 157 |
+
"rewards/margins": 0.02389874868094921,
|
| 158 |
+
"rewards/real": -0.010342543944716454,
|
| 159 |
+
"step": 45
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.06,
|
| 163 |
+
"grad_norm": 26.875,
|
| 164 |
+
"learning_rate": 3.1645569620253163e-07,
|
| 165 |
+
"logits/generated": 0.362557053565979,
|
| 166 |
+
"logits/real": -0.13479682803153992,
|
| 167 |
+
"logps/generated": -559.6375122070312,
|
| 168 |
+
"logps/real": -434.8016662597656,
|
| 169 |
+
"loss": 0.4702,
|
| 170 |
+
"rewards/accuracies": 0.59375,
|
| 171 |
+
"rewards/generated": -0.0708869993686676,
|
| 172 |
+
"rewards/margins": 0.08817923069000244,
|
| 173 |
+
"rewards/real": 0.01729224994778633,
|
| 174 |
+
"step": 50
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.07,
|
| 178 |
+
"grad_norm": 33.25,
|
| 179 |
+
"learning_rate": 3.481012658227848e-07,
|
| 180 |
+
"logits/generated": 0.3691774010658264,
|
| 181 |
+
"logits/real": -0.11431871354579926,
|
| 182 |
+
"logps/generated": -591.8275146484375,
|
| 183 |
+
"logps/real": -450.1565856933594,
|
| 184 |
+
"loss": 0.4806,
|
| 185 |
+
"rewards/accuracies": 0.5625,
|
| 186 |
+
"rewards/generated": -0.07000661641359329,
|
| 187 |
+
"rewards/margins": 0.05859777331352234,
|
| 188 |
+
"rewards/real": -0.011408851481974125,
|
| 189 |
+
"step": 55
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.08,
|
| 193 |
+
"grad_norm": 32.0,
|
| 194 |
+
"learning_rate": 3.79746835443038e-07,
|
| 195 |
+
"logits/generated": 0.3952616751194,
|
| 196 |
+
"logits/real": -0.08973407000303268,
|
| 197 |
+
"logps/generated": -569.9074096679688,
|
| 198 |
+
"logps/real": -456.4561462402344,
|
| 199 |
+
"loss": 0.4661,
|
| 200 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 201 |
+
"rewards/generated": -0.06687521934509277,
|
| 202 |
+
"rewards/margins": 0.08400523662567139,
|
| 203 |
+
"rewards/real": 0.017130006104707718,
|
| 204 |
+
"step": 60
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.08,
|
| 208 |
+
"grad_norm": 27.25,
|
| 209 |
+
"learning_rate": 4.1139240506329117e-07,
|
| 210 |
+
"logits/generated": 0.36575227975845337,
|
| 211 |
+
"logits/real": -0.1133301854133606,
|
| 212 |
+
"logps/generated": -542.1326293945312,
|
| 213 |
+
"logps/real": -441.4345703125,
|
| 214 |
+
"loss": 0.4689,
|
| 215 |
+
"rewards/accuracies": 0.59375,
|
| 216 |
+
"rewards/generated": -0.07646136730909348,
|
| 217 |
+
"rewards/margins": 0.08980376273393631,
|
| 218 |
+
"rewards/real": 0.013342398218810558,
|
| 219 |
+
"step": 65
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09,
|
| 223 |
+
"grad_norm": 29.875,
|
| 224 |
+
"learning_rate": 4.4303797468354424e-07,
|
| 225 |
+
"logits/generated": 0.3912655711174011,
|
| 226 |
+
"logits/real": -0.1571967601776123,
|
| 227 |
+
"logps/generated": -570.942626953125,
|
| 228 |
+
"logps/real": -486.6839904785156,
|
| 229 |
+
"loss": 0.4531,
|
| 230 |
+
"rewards/accuracies": 0.668749988079071,
|
| 231 |
+
"rewards/generated": -0.13936015963554382,
|
| 232 |
+
"rewards/margins": 0.1767238825559616,
|
| 233 |
+
"rewards/real": 0.03736373037099838,
|
| 234 |
+
"step": 70
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1,
|
| 238 |
+
"grad_norm": 26.5,
|
| 239 |
+
"learning_rate": 4.746835443037974e-07,
|
| 240 |
+
"logits/generated": 0.3816757798194885,
|
| 241 |
+
"logits/real": -0.09128452837467194,
|
| 242 |
+
"logps/generated": -571.598876953125,
|
| 243 |
+
"logps/real": -419.603515625,
|
| 244 |
+
"loss": 0.4616,
|
| 245 |
+
"rewards/accuracies": 0.606249988079071,
|
| 246 |
+
"rewards/generated": -0.09458805620670319,
|
| 247 |
+
"rewards/margins": 0.08172789961099625,
|
| 248 |
+
"rewards/real": -0.012860171496868134,
|
| 249 |
+
"step": 75
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.1,
|
| 253 |
+
"grad_norm": 28.375,
|
| 254 |
+
"learning_rate": 4.992877492877492e-07,
|
| 255 |
+
"logits/generated": 0.34975099563598633,
|
| 256 |
+
"logits/real": -0.14952535927295685,
|
| 257 |
+
"logps/generated": -547.2381591796875,
|
| 258 |
+
"logps/real": -442.2696838378906,
|
| 259 |
+
"loss": 0.4441,
|
| 260 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 261 |
+
"rewards/generated": -0.13002917170524597,
|
| 262 |
+
"rewards/margins": 0.15946698188781738,
|
| 263 |
+
"rewards/real": 0.029437804594635963,
|
| 264 |
+
"step": 80
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.11,
|
| 268 |
+
"grad_norm": 28.75,
|
| 269 |
+
"learning_rate": 4.957264957264958e-07,
|
| 270 |
+
"logits/generated": 0.35858625173568726,
|
| 271 |
+
"logits/real": -0.18670549988746643,
|
| 272 |
+
"logps/generated": -535.11669921875,
|
| 273 |
+
"logps/real": -456.5271911621094,
|
| 274 |
+
"loss": 0.4408,
|
| 275 |
+
"rewards/accuracies": 0.737500011920929,
|
| 276 |
+
"rewards/generated": -0.18137022852897644,
|
| 277 |
+
"rewards/margins": 0.22222211956977844,
|
| 278 |
+
"rewards/real": 0.040851861238479614,
|
| 279 |
+
"step": 85
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.12,
|
| 283 |
+
"grad_norm": 29.5,
|
| 284 |
+
"learning_rate": 4.921652421652421e-07,
|
| 285 |
+
"logits/generated": 0.30655255913734436,
|
| 286 |
+
"logits/real": -0.13807684183120728,
|
| 287 |
+
"logps/generated": -579.0311279296875,
|
| 288 |
+
"logps/real": -435.41485595703125,
|
| 289 |
+
"loss": 0.4332,
|
| 290 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 291 |
+
"rewards/generated": -0.17220628261566162,
|
| 292 |
+
"rewards/margins": 0.18943142890930176,
|
| 293 |
+
"rewards/real": 0.01722514070570469,
|
| 294 |
+
"step": 90
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.12,
|
| 298 |
+
"grad_norm": 28.25,
|
| 299 |
+
"learning_rate": 4.886039886039886e-07,
|
| 300 |
+
"logits/generated": 0.3390645980834961,
|
| 301 |
+
"logits/real": -0.159017413854599,
|
| 302 |
+
"logps/generated": -562.1036376953125,
|
| 303 |
+
"logps/real": -472.2007751464844,
|
| 304 |
+
"loss": 0.4335,
|
| 305 |
+
"rewards/accuracies": 0.706250011920929,
|
| 306 |
+
"rewards/generated": -0.19639980792999268,
|
| 307 |
+
"rewards/margins": 0.21172475814819336,
|
| 308 |
+
"rewards/real": 0.015324937179684639,
|
| 309 |
+
"step": 95
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.13,
|
| 313 |
+
"grad_norm": 27.375,
|
| 314 |
+
"learning_rate": 4.850427350427351e-07,
|
| 315 |
+
"logits/generated": 0.3581392168998718,
|
| 316 |
+
"logits/real": -0.22140832245349884,
|
| 317 |
+
"logps/generated": -518.3876953125,
|
| 318 |
+
"logps/real": -505.3263244628906,
|
| 319 |
+
"loss": 0.4226,
|
| 320 |
+
"rewards/accuracies": 0.731249988079071,
|
| 321 |
+
"rewards/generated": -0.22456979751586914,
|
| 322 |
+
"rewards/margins": 0.27873098850250244,
|
| 323 |
+
"rewards/real": 0.0541611909866333,
|
| 324 |
+
"step": 100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.13,
|
| 328 |
+
"grad_norm": 29.0,
|
| 329 |
+
"learning_rate": 4.814814814814814e-07,
|
| 330 |
+
"logits/generated": 0.3360639214515686,
|
| 331 |
+
"logits/real": -0.16623374819755554,
|
| 332 |
+
"logps/generated": -597.8626708984375,
|
| 333 |
+
"logps/real": -469.96435546875,
|
| 334 |
+
"loss": 0.4089,
|
| 335 |
+
"rewards/accuracies": 0.824999988079071,
|
| 336 |
+
"rewards/generated": -0.2624703645706177,
|
| 337 |
+
"rewards/margins": 0.3305851221084595,
|
| 338 |
+
"rewards/real": 0.06811471283435822,
|
| 339 |
+
"step": 105
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.14,
|
| 343 |
+
"grad_norm": 27.125,
|
| 344 |
+
"learning_rate": 4.779202279202279e-07,
|
| 345 |
+
"logits/generated": 0.3120338022708893,
|
| 346 |
+
"logits/real": -0.1398223340511322,
|
| 347 |
+
"logps/generated": -559.6121826171875,
|
| 348 |
+
"logps/real": -435.9784240722656,
|
| 349 |
+
"loss": 0.4151,
|
| 350 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 351 |
+
"rewards/generated": -0.2531023323535919,
|
| 352 |
+
"rewards/margins": 0.2851903438568115,
|
| 353 |
+
"rewards/real": 0.03208799287676811,
|
| 354 |
+
"step": 110
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.15,
|
| 358 |
+
"grad_norm": 27.0,
|
| 359 |
+
"learning_rate": 4.743589743589743e-07,
|
| 360 |
+
"logits/generated": 0.35134896636009216,
|
| 361 |
+
"logits/real": -0.16236919164657593,
|
| 362 |
+
"logps/generated": -540.6229248046875,
|
| 363 |
+
"logps/real": -481.10479736328125,
|
| 364 |
+
"loss": 0.3949,
|
| 365 |
+
"rewards/accuracies": 0.768750011920929,
|
| 366 |
+
"rewards/generated": -0.3117538094520569,
|
| 367 |
+
"rewards/margins": 0.37438851594924927,
|
| 368 |
+
"rewards/real": 0.06263472884893417,
|
| 369 |
+
"step": 115
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.15,
|
| 373 |
+
"grad_norm": 28.0,
|
| 374 |
+
"learning_rate": 4.707977207977208e-07,
|
| 375 |
+
"logits/generated": 0.31295281648635864,
|
| 376 |
+
"logits/real": -0.15100577473640442,
|
| 377 |
+
"logps/generated": -533.11376953125,
|
| 378 |
+
"logps/real": -425.769775390625,
|
| 379 |
+
"loss": 0.4001,
|
| 380 |
+
"rewards/accuracies": 0.856249988079071,
|
| 381 |
+
"rewards/generated": -0.3296114504337311,
|
| 382 |
+
"rewards/margins": 0.37464508414268494,
|
| 383 |
+
"rewards/real": 0.04503365978598595,
|
| 384 |
+
"step": 120
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.16,
|
| 388 |
+
"grad_norm": 28.375,
|
| 389 |
+
"learning_rate": 4.672364672364672e-07,
|
| 390 |
+
"logits/generated": 0.3123488128185272,
|
| 391 |
+
"logits/real": -0.1613120287656784,
|
| 392 |
+
"logps/generated": -556.4259643554688,
|
| 393 |
+
"logps/real": -451.27593994140625,
|
| 394 |
+
"loss": 0.3915,
|
| 395 |
+
"rewards/accuracies": 0.824999988079071,
|
| 396 |
+
"rewards/generated": -0.32520076632499695,
|
| 397 |
+
"rewards/margins": 0.3783460259437561,
|
| 398 |
+
"rewards/real": 0.05314524844288826,
|
| 399 |
+
"step": 125
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.17,
|
| 403 |
+
"grad_norm": 28.875,
|
| 404 |
+
"learning_rate": 4.6367521367521367e-07,
|
| 405 |
+
"logits/generated": 0.3258894085884094,
|
| 406 |
+
"logits/real": -0.1628173589706421,
|
| 407 |
+
"logps/generated": -558.4219360351562,
|
| 408 |
+
"logps/real": -435.81243896484375,
|
| 409 |
+
"loss": 0.3934,
|
| 410 |
+
"rewards/accuracies": 0.8687499761581421,
|
| 411 |
+
"rewards/generated": -0.36564749479293823,
|
| 412 |
+
"rewards/margins": 0.4104110300540924,
|
| 413 |
+
"rewards/real": 0.04476350545883179,
|
| 414 |
+
"step": 130
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.17,
|
| 418 |
+
"grad_norm": 26.75,
|
| 419 |
+
"learning_rate": 4.601139601139601e-07,
|
| 420 |
+
"logits/generated": 0.2820456326007843,
|
| 421 |
+
"logits/real": -0.2306266725063324,
|
| 422 |
+
"logps/generated": -578.9539794921875,
|
| 423 |
+
"logps/real": -447.5259704589844,
|
| 424 |
+
"loss": 0.3839,
|
| 425 |
+
"rewards/accuracies": 0.824999988079071,
|
| 426 |
+
"rewards/generated": -0.3908230662345886,
|
| 427 |
+
"rewards/margins": 0.4163753390312195,
|
| 428 |
+
"rewards/real": 0.02555229887366295,
|
| 429 |
+
"step": 135
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.18,
|
| 433 |
+
"grad_norm": 28.25,
|
| 434 |
+
"learning_rate": 4.5655270655270654e-07,
|
| 435 |
+
"logits/generated": 0.3293055295944214,
|
| 436 |
+
"logits/real": -0.24442501366138458,
|
| 437 |
+
"logps/generated": -557.1941528320312,
|
| 438 |
+
"logps/real": -533.494873046875,
|
| 439 |
+
"loss": 0.3766,
|
| 440 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 441 |
+
"rewards/generated": -0.37160658836364746,
|
| 442 |
+
"rewards/margins": 0.4162415862083435,
|
| 443 |
+
"rewards/real": 0.044634997844696045,
|
| 444 |
+
"step": 140
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.19,
|
| 448 |
+
"grad_norm": 28.0,
|
| 449 |
+
"learning_rate": 4.5299145299145297e-07,
|
| 450 |
+
"logits/generated": 0.28380945324897766,
|
| 451 |
+
"logits/real": -0.16644850373268127,
|
| 452 |
+
"logps/generated": -532.15625,
|
| 453 |
+
"logps/real": -427.69287109375,
|
| 454 |
+
"loss": 0.3824,
|
| 455 |
+
"rewards/accuracies": 0.84375,
|
| 456 |
+
"rewards/generated": -0.42885756492614746,
|
| 457 |
+
"rewards/margins": 0.4583609104156494,
|
| 458 |
+
"rewards/real": 0.029503341764211655,
|
| 459 |
+
"step": 145
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.19,
|
| 463 |
+
"grad_norm": 26.375,
|
| 464 |
+
"learning_rate": 4.494301994301994e-07,
|
| 465 |
+
"logits/generated": 0.26398712396621704,
|
| 466 |
+
"logits/real": -0.1775752604007721,
|
| 467 |
+
"logps/generated": -544.0301513671875,
|
| 468 |
+
"logps/real": -442.7901916503906,
|
| 469 |
+
"loss": 0.3883,
|
| 470 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 471 |
+
"rewards/generated": -0.4080891013145447,
|
| 472 |
+
"rewards/margins": 0.4677355885505676,
|
| 473 |
+
"rewards/real": 0.05964647978544235,
|
| 474 |
+
"step": 150
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.2,
|
| 478 |
+
"grad_norm": 28.125,
|
| 479 |
+
"learning_rate": 4.4586894586894584e-07,
|
| 480 |
+
"logits/generated": 0.3235207796096802,
|
| 481 |
+
"logits/real": -0.19184669852256775,
|
| 482 |
+
"logps/generated": -565.9249267578125,
|
| 483 |
+
"logps/real": -485.18109130859375,
|
| 484 |
+
"loss": 0.3713,
|
| 485 |
+
"rewards/accuracies": 0.862500011920929,
|
| 486 |
+
"rewards/generated": -0.43480604887008667,
|
| 487 |
+
"rewards/margins": 0.4802734851837158,
|
| 488 |
+
"rewards/real": 0.04546744376420975,
|
| 489 |
+
"step": 155
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.2,
|
| 493 |
+
"grad_norm": 27.875,
|
| 494 |
+
"learning_rate": 4.423076923076923e-07,
|
| 495 |
+
"logits/generated": 0.3015090823173523,
|
| 496 |
+
"logits/real": -0.2252935916185379,
|
| 497 |
+
"logps/generated": -566.9320068359375,
|
| 498 |
+
"logps/real": -444.4369201660156,
|
| 499 |
+
"loss": 0.3552,
|
| 500 |
+
"rewards/accuracies": 0.8687499761581421,
|
| 501 |
+
"rewards/generated": -0.45722684264183044,
|
| 502 |
+
"rewards/margins": 0.554154634475708,
|
| 503 |
+
"rewards/real": 0.09692778438329697,
|
| 504 |
+
"step": 160
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.21,
|
| 508 |
+
"grad_norm": 26.5,
|
| 509 |
+
"learning_rate": 4.3874643874643876e-07,
|
| 510 |
+
"logits/generated": 0.31183922290802,
|
| 511 |
+
"logits/real": -0.1933097243309021,
|
| 512 |
+
"logps/generated": -558.2478637695312,
|
| 513 |
+
"logps/real": -460.18267822265625,
|
| 514 |
+
"loss": 0.3597,
|
| 515 |
+
"rewards/accuracies": 0.831250011920929,
|
| 516 |
+
"rewards/generated": -0.47245973348617554,
|
| 517 |
+
"rewards/margins": 0.5111454129219055,
|
| 518 |
+
"rewards/real": 0.03868564963340759,
|
| 519 |
+
"step": 165
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.22,
|
| 523 |
+
"grad_norm": 27.375,
|
| 524 |
+
"learning_rate": 4.3518518518518514e-07,
|
| 525 |
+
"logits/generated": 0.250281423330307,
|
| 526 |
+
"logits/real": -0.1833321750164032,
|
| 527 |
+
"logps/generated": -536.4636840820312,
|
| 528 |
+
"logps/real": -416.641357421875,
|
| 529 |
+
"loss": 0.364,
|
| 530 |
+
"rewards/accuracies": 0.856249988079071,
|
| 531 |
+
"rewards/generated": -0.454394668340683,
|
| 532 |
+
"rewards/margins": 0.5187661647796631,
|
| 533 |
+
"rewards/real": 0.06437145173549652,
|
| 534 |
+
"step": 170
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22,
|
| 538 |
+
"grad_norm": 26.625,
|
| 539 |
+
"learning_rate": 4.3162393162393163e-07,
|
| 540 |
+
"logits/generated": 0.29415473341941833,
|
| 541 |
+
"logits/real": -0.24404272437095642,
|
| 542 |
+
"logps/generated": -538.131103515625,
|
| 543 |
+
"logps/real": -477.87603759765625,
|
| 544 |
+
"loss": 0.348,
|
| 545 |
+
"rewards/accuracies": 0.925000011920929,
|
| 546 |
+
"rewards/generated": -0.5133287310600281,
|
| 547 |
+
"rewards/margins": 0.5939925909042358,
|
| 548 |
+
"rewards/real": 0.08066388964653015,
|
| 549 |
+
"step": 175
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.23,
|
| 553 |
+
"grad_norm": 30.125,
|
| 554 |
+
"learning_rate": 4.2806267806267807e-07,
|
| 555 |
+
"logits/generated": 0.27652794122695923,
|
| 556 |
+
"logits/real": -0.2590489685535431,
|
| 557 |
+
"logps/generated": -561.6041259765625,
|
| 558 |
+
"logps/real": -444.3932189941406,
|
| 559 |
+
"loss": 0.3536,
|
| 560 |
+
"rewards/accuracies": 0.887499988079071,
|
| 561 |
+
"rewards/generated": -0.47919169068336487,
|
| 562 |
+
"rewards/margins": 0.5749484896659851,
|
| 563 |
+
"rewards/real": 0.09575681388378143,
|
| 564 |
+
"step": 180
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.24,
|
| 568 |
+
"grad_norm": 26.75,
|
| 569 |
+
"learning_rate": 4.245014245014245e-07,
|
| 570 |
+
"logits/generated": 0.3380230963230133,
|
| 571 |
+
"logits/real": -0.11477069556713104,
|
| 572 |
+
"logps/generated": -599.4484252929688,
|
| 573 |
+
"logps/real": -421.1564025878906,
|
| 574 |
+
"loss": 0.3357,
|
| 575 |
+
"rewards/accuracies": 0.925000011920929,
|
| 576 |
+
"rewards/generated": -0.5778478384017944,
|
| 577 |
+
"rewards/margins": 0.6686479449272156,
|
| 578 |
+
"rewards/real": 0.09080011397600174,
|
| 579 |
+
"step": 185
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.24,
|
| 583 |
+
"grad_norm": 27.0,
|
| 584 |
+
"learning_rate": 4.2094017094017093e-07,
|
| 585 |
+
"logits/generated": 0.3196483850479126,
|
| 586 |
+
"logits/real": -0.1744014322757721,
|
| 587 |
+
"logps/generated": -586.2642822265625,
|
| 588 |
+
"logps/real": -426.1298828125,
|
| 589 |
+
"loss": 0.3573,
|
| 590 |
+
"rewards/accuracies": 0.875,
|
| 591 |
+
"rewards/generated": -0.48544687032699585,
|
| 592 |
+
"rewards/margins": 0.557063102722168,
|
| 593 |
+
"rewards/real": 0.07161618769168854,
|
| 594 |
+
"step": 190
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.25,
|
| 598 |
+
"grad_norm": 27.125,
|
| 599 |
+
"learning_rate": 4.173789173789173e-07,
|
| 600 |
+
"logits/generated": 0.25956088304519653,
|
| 601 |
+
"logits/real": -0.2154865711927414,
|
| 602 |
+
"logps/generated": -528.0460205078125,
|
| 603 |
+
"logps/real": -452.2561950683594,
|
| 604 |
+
"loss": 0.3429,
|
| 605 |
+
"rewards/accuracies": 0.90625,
|
| 606 |
+
"rewards/generated": -0.5490601658821106,
|
| 607 |
+
"rewards/margins": 0.6324241757392883,
|
| 608 |
+
"rewards/real": 0.08336395770311356,
|
| 609 |
+
"step": 195
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.26,
|
| 613 |
+
"grad_norm": 26.5,
|
| 614 |
+
"learning_rate": 4.138176638176638e-07,
|
| 615 |
+
"logits/generated": 0.3287113904953003,
|
| 616 |
+
"logits/real": -0.20184390246868134,
|
| 617 |
+
"logps/generated": -580.03857421875,
|
| 618 |
+
"logps/real": -471.832763671875,
|
| 619 |
+
"loss": 0.3437,
|
| 620 |
+
"rewards/accuracies": 0.875,
|
| 621 |
+
"rewards/generated": -0.5396500825881958,
|
| 622 |
+
"rewards/margins": 0.575441837310791,
|
| 623 |
+
"rewards/real": 0.03579169511795044,
|
| 624 |
+
"step": 200
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.26,
|
| 628 |
+
"grad_norm": 25.5,
|
| 629 |
+
"learning_rate": 4.1025641025641024e-07,
|
| 630 |
+
"logits/generated": 0.3303323984146118,
|
| 631 |
+
"logits/real": -0.18564388155937195,
|
| 632 |
+
"logps/generated": -576.2296752929688,
|
| 633 |
+
"logps/real": -470.49554443359375,
|
| 634 |
+
"loss": 0.3435,
|
| 635 |
+
"rewards/accuracies": 0.856249988079071,
|
| 636 |
+
"rewards/generated": -0.5153689384460449,
|
| 637 |
+
"rewards/margins": 0.5805807113647461,
|
| 638 |
+
"rewards/real": 0.06521172821521759,
|
| 639 |
+
"step": 205
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.27,
|
| 643 |
+
"grad_norm": 28.0,
|
| 644 |
+
"learning_rate": 4.0669515669515667e-07,
|
| 645 |
+
"logits/generated": 0.25469496846199036,
|
| 646 |
+
"logits/real": -0.28663891553878784,
|
| 647 |
+
"logps/generated": -558.517822265625,
|
| 648 |
+
"logps/real": -467.5023498535156,
|
| 649 |
+
"loss": 0.3327,
|
| 650 |
+
"rewards/accuracies": 0.875,
|
| 651 |
+
"rewards/generated": -0.5688196420669556,
|
| 652 |
+
"rewards/margins": 0.6551179885864258,
|
| 653 |
+
"rewards/real": 0.08629832416772842,
|
| 654 |
+
"step": 210
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.28,
|
| 658 |
+
"grad_norm": 23.75,
|
| 659 |
+
"learning_rate": 4.031339031339031e-07,
|
| 660 |
+
"logits/generated": 0.2868548333644867,
|
| 661 |
+
"logits/real": -0.19922226667404175,
|
| 662 |
+
"logps/generated": -549.5493774414062,
|
| 663 |
+
"logps/real": -457.3771057128906,
|
| 664 |
+
"loss": 0.3296,
|
| 665 |
+
"rewards/accuracies": 0.8812500238418579,
|
| 666 |
+
"rewards/generated": -0.6514773964881897,
|
| 667 |
+
"rewards/margins": 0.7027413845062256,
|
| 668 |
+
"rewards/real": 0.05126393958926201,
|
| 669 |
+
"step": 215
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.28,
|
| 673 |
+
"grad_norm": 25.375,
|
| 674 |
+
"learning_rate": 3.995726495726496e-07,
|
| 675 |
+
"logits/generated": 0.3001692295074463,
|
| 676 |
+
"logits/real": -0.18475447595119476,
|
| 677 |
+
"logps/generated": -541.4611206054688,
|
| 678 |
+
"logps/real": -444.6468200683594,
|
| 679 |
+
"loss": 0.338,
|
| 680 |
+
"rewards/accuracies": 0.856249988079071,
|
| 681 |
+
"rewards/generated": -0.5863200426101685,
|
| 682 |
+
"rewards/margins": 0.6246292591094971,
|
| 683 |
+
"rewards/real": 0.03830911964178085,
|
| 684 |
+
"step": 220
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.29,
|
| 688 |
+
"grad_norm": 24.625,
|
| 689 |
+
"learning_rate": 3.9601139601139597e-07,
|
| 690 |
+
"logits/generated": 0.2538192868232727,
|
| 691 |
+
"logits/real": -0.2353261411190033,
|
| 692 |
+
"logps/generated": -573.8902587890625,
|
| 693 |
+
"logps/real": -450.2127380371094,
|
| 694 |
+
"loss": 0.3309,
|
| 695 |
+
"rewards/accuracies": 0.8999999761581421,
|
| 696 |
+
"rewards/generated": -0.5875332951545715,
|
| 697 |
+
"rewards/margins": 0.6803909540176392,
|
| 698 |
+
"rewards/real": 0.09285767376422882,
|
| 699 |
+
"step": 225
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.29,
|
| 703 |
+
"grad_norm": 26.125,
|
| 704 |
+
"learning_rate": 3.924501424501424e-07,
|
| 705 |
+
"logits/generated": 0.2551348805427551,
|
| 706 |
+
"logits/real": -0.2044079303741455,
|
| 707 |
+
"logps/generated": -546.6871337890625,
|
| 708 |
+
"logps/real": -421.7835998535156,
|
| 709 |
+
"loss": 0.3181,
|
| 710 |
+
"rewards/accuracies": 0.893750011920929,
|
| 711 |
+
"rewards/generated": -0.5918381214141846,
|
| 712 |
+
"rewards/margins": 0.6859561204910278,
|
| 713 |
+
"rewards/real": 0.09411804378032684,
|
| 714 |
+
"step": 230
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.3,
|
| 718 |
+
"grad_norm": 25.375,
|
| 719 |
+
"learning_rate": 3.888888888888889e-07,
|
| 720 |
+
"logits/generated": 0.3060089349746704,
|
| 721 |
+
"logits/real": -0.21332141757011414,
|
| 722 |
+
"logps/generated": -597.86376953125,
|
| 723 |
+
"logps/real": -443.6407165527344,
|
| 724 |
+
"loss": 0.32,
|
| 725 |
+
"rewards/accuracies": 0.90625,
|
| 726 |
+
"rewards/generated": -0.6655870676040649,
|
| 727 |
+
"rewards/margins": 0.729436993598938,
|
| 728 |
+
"rewards/real": 0.06384985148906708,
|
| 729 |
+
"step": 235
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.31,
|
| 733 |
+
"grad_norm": 26.125,
|
| 734 |
+
"learning_rate": 3.853276353276353e-07,
|
| 735 |
+
"logits/generated": 0.30169203877449036,
|
| 736 |
+
"logits/real": -0.1919897347688675,
|
| 737 |
+
"logps/generated": -541.0394287109375,
|
| 738 |
+
"logps/real": -423.79833984375,
|
| 739 |
+
"loss": 0.3223,
|
| 740 |
+
"rewards/accuracies": 0.918749988079071,
|
| 741 |
+
"rewards/generated": -0.5991157293319702,
|
| 742 |
+
"rewards/margins": 0.68681800365448,
|
| 743 |
+
"rewards/real": 0.08770229667425156,
|
| 744 |
+
"step": 240
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.31,
|
| 748 |
+
"grad_norm": 25.375,
|
| 749 |
+
"learning_rate": 3.8176638176638176e-07,
|
| 750 |
+
"logits/generated": 0.2983294129371643,
|
| 751 |
+
"logits/real": -0.23591478168964386,
|
| 752 |
+
"logps/generated": -559.3455200195312,
|
| 753 |
+
"logps/real": -503.95751953125,
|
| 754 |
+
"loss": 0.3408,
|
| 755 |
+
"rewards/accuracies": 0.8812500238418579,
|
| 756 |
+
"rewards/generated": -0.6077674031257629,
|
| 757 |
+
"rewards/margins": 0.6536551713943481,
|
| 758 |
+
"rewards/real": 0.04588776454329491,
|
| 759 |
+
"step": 245
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.32,
|
| 763 |
+
"grad_norm": 27.125,
|
| 764 |
+
"learning_rate": 3.782051282051282e-07,
|
| 765 |
+
"logits/generated": 0.26214712858200073,
|
| 766 |
+
"logits/real": -0.2555038034915924,
|
| 767 |
+
"logps/generated": -555.8446044921875,
|
| 768 |
+
"logps/real": -438.34423828125,
|
| 769 |
+
"loss": 0.3161,
|
| 770 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 771 |
+
"rewards/generated": -0.6663080453872681,
|
| 772 |
+
"rewards/margins": 0.738936185836792,
|
| 773 |
+
"rewards/real": 0.07262810319662094,
|
| 774 |
+
"step": 250
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.33,
|
| 778 |
+
"grad_norm": 24.75,
|
| 779 |
+
"learning_rate": 3.7464387464387463e-07,
|
| 780 |
+
"logits/generated": 0.30381038784980774,
|
| 781 |
+
"logits/real": -0.2225998193025589,
|
| 782 |
+
"logps/generated": -538.4495849609375,
|
| 783 |
+
"logps/real": -484.36993408203125,
|
| 784 |
+
"loss": 0.3231,
|
| 785 |
+
"rewards/accuracies": 0.9125000238418579,
|
| 786 |
+
"rewards/generated": -0.6721928715705872,
|
| 787 |
+
"rewards/margins": 0.7017067670822144,
|
| 788 |
+
"rewards/real": 0.02951394021511078,
|
| 789 |
+
"step": 255
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.33,
|
| 793 |
+
"grad_norm": 25.5,
|
| 794 |
+
"learning_rate": 3.7108262108262107e-07,
|
| 795 |
+
"logits/generated": 0.28403228521347046,
|
| 796 |
+
"logits/real": -0.24081599712371826,
|
| 797 |
+
"logps/generated": -552.0731811523438,
|
| 798 |
+
"logps/real": -438.84356689453125,
|
| 799 |
+
"loss": 0.3093,
|
| 800 |
+
"rewards/accuracies": 0.887499988079071,
|
| 801 |
+
"rewards/generated": -0.6770501732826233,
|
| 802 |
+
"rewards/margins": 0.7596859335899353,
|
| 803 |
+
"rewards/real": 0.08263583481311798,
|
| 804 |
+
"step": 260
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.34,
|
| 808 |
+
"grad_norm": 26.0,
|
| 809 |
+
"learning_rate": 3.6752136752136755e-07,
|
| 810 |
+
"logits/generated": 0.2751469016075134,
|
| 811 |
+
"logits/real": -0.2667590081691742,
|
| 812 |
+
"logps/generated": -574.9193115234375,
|
| 813 |
+
"logps/real": -458.193359375,
|
| 814 |
+
"loss": 0.3005,
|
| 815 |
+
"rewards/accuracies": 0.96875,
|
| 816 |
+
"rewards/generated": -0.6883004307746887,
|
| 817 |
+
"rewards/margins": 0.8165790438652039,
|
| 818 |
+
"rewards/real": 0.1282786875963211,
|
| 819 |
+
"step": 265
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.35,
|
| 823 |
+
"grad_norm": 28.125,
|
| 824 |
+
"learning_rate": 3.6396011396011393e-07,
|
| 825 |
+
"logits/generated": 0.3092981278896332,
|
| 826 |
+
"logits/real": -0.24193081259727478,
|
| 827 |
+
"logps/generated": -537.2657470703125,
|
| 828 |
+
"logps/real": -463.10784912109375,
|
| 829 |
+
"loss": 0.3048,
|
| 830 |
+
"rewards/accuracies": 0.918749988079071,
|
| 831 |
+
"rewards/generated": -0.6616233587265015,
|
| 832 |
+
"rewards/margins": 0.76463383436203,
|
| 833 |
+
"rewards/real": 0.10301046073436737,
|
| 834 |
+
"step": 270
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.35,
|
| 838 |
+
"grad_norm": 25.0,
|
| 839 |
+
"learning_rate": 3.6039886039886037e-07,
|
| 840 |
+
"logits/generated": 0.29445740580558777,
|
| 841 |
+
"logits/real": -0.21788445115089417,
|
| 842 |
+
"logps/generated": -562.7876586914062,
|
| 843 |
+
"logps/real": -432.4056701660156,
|
| 844 |
+
"loss": 0.2912,
|
| 845 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 846 |
+
"rewards/generated": -0.7921088337898254,
|
| 847 |
+
"rewards/margins": 0.866392970085144,
|
| 848 |
+
"rewards/real": 0.07428419589996338,
|
| 849 |
+
"step": 275
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.36,
|
| 853 |
+
"grad_norm": 28.125,
|
| 854 |
+
"learning_rate": 3.5683760683760686e-07,
|
| 855 |
+
"logits/generated": 0.29365968704223633,
|
| 856 |
+
"logits/real": -0.16456809639930725,
|
| 857 |
+
"logps/generated": -568.531005859375,
|
| 858 |
+
"logps/real": -452.4728088378906,
|
| 859 |
+
"loss": 0.2857,
|
| 860 |
+
"rewards/accuracies": 0.925000011920929,
|
| 861 |
+
"rewards/generated": -0.7912765145301819,
|
| 862 |
+
"rewards/margins": 0.9027554392814636,
|
| 863 |
+
"rewards/real": 0.11147890239953995,
|
| 864 |
+
"step": 280
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.36,
|
| 868 |
+
"grad_norm": 29.875,
|
| 869 |
+
"learning_rate": 3.5327635327635324e-07,
|
| 870 |
+
"logits/generated": 0.2611847519874573,
|
| 871 |
+
"logits/real": -0.23247161507606506,
|
| 872 |
+
"logps/generated": -557.5487060546875,
|
| 873 |
+
"logps/real": -437.1328125,
|
| 874 |
+
"loss": 0.301,
|
| 875 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 876 |
+
"rewards/generated": -0.7435336112976074,
|
| 877 |
+
"rewards/margins": 0.8042283058166504,
|
| 878 |
+
"rewards/real": 0.0606946162879467,
|
| 879 |
+
"step": 285
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.37,
|
| 883 |
+
"grad_norm": 26.5,
|
| 884 |
+
"learning_rate": 3.497150997150997e-07,
|
| 885 |
+
"logits/generated": 0.2905731797218323,
|
| 886 |
+
"logits/real": -0.29508811235427856,
|
| 887 |
+
"logps/generated": -568.8480834960938,
|
| 888 |
+
"logps/real": -493.10662841796875,
|
| 889 |
+
"loss": 0.298,
|
| 890 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 891 |
+
"rewards/generated": -0.8001793622970581,
|
| 892 |
+
"rewards/margins": 0.8642541766166687,
|
| 893 |
+
"rewards/real": 0.06407480686903,
|
| 894 |
+
"step": 290
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.38,
|
| 898 |
+
"grad_norm": 25.125,
|
| 899 |
+
"learning_rate": 3.461538461538461e-07,
|
| 900 |
+
"logits/generated": 0.25090640783309937,
|
| 901 |
+
"logits/real": -0.23970182240009308,
|
| 902 |
+
"logps/generated": -571.603759765625,
|
| 903 |
+
"logps/real": -419.73162841796875,
|
| 904 |
+
"loss": 0.289,
|
| 905 |
+
"rewards/accuracies": 0.949999988079071,
|
| 906 |
+
"rewards/generated": -0.8356046676635742,
|
| 907 |
+
"rewards/margins": 0.9479316473007202,
|
| 908 |
+
"rewards/real": 0.11232707649469376,
|
| 909 |
+
"step": 295
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.38,
|
| 913 |
+
"grad_norm": 23.0,
|
| 914 |
+
"learning_rate": 3.425925925925926e-07,
|
| 915 |
+
"logits/generated": 0.24801941215991974,
|
| 916 |
+
"logits/real": -0.2514159381389618,
|
| 917 |
+
"logps/generated": -578.6694946289062,
|
| 918 |
+
"logps/real": -446.20550537109375,
|
| 919 |
+
"loss": 0.2932,
|
| 920 |
+
"rewards/accuracies": 0.887499988079071,
|
| 921 |
+
"rewards/generated": -0.7759395837783813,
|
| 922 |
+
"rewards/margins": 0.8407967686653137,
|
| 923 |
+
"rewards/real": 0.06485719978809357,
|
| 924 |
+
"step": 300
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.38,
|
| 928 |
+
"eval_logits/generated": 0.25535792112350464,
|
| 929 |
+
"eval_logits/real": -0.2529524266719818,
|
| 930 |
+
"eval_logps/generated": -568.39892578125,
|
| 931 |
+
"eval_logps/real": -468.0555725097656,
|
| 932 |
+
"eval_loss": 0.29618778824806213,
|
| 933 |
+
"eval_rewards/accuracies": 0.921999990940094,
|
| 934 |
+
"eval_rewards/generated": -0.7854698300361633,
|
| 935 |
+
"eval_rewards/margins": 0.857230007648468,
|
| 936 |
+
"eval_rewards/real": 0.07176008075475693,
|
| 937 |
+
"eval_runtime": 251.8666,
|
| 938 |
+
"eval_samples_per_second": 7.941,
|
| 939 |
+
"eval_steps_per_second": 0.993,
|
| 940 |
+
"step": 300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.39,
|
| 944 |
+
"grad_norm": 27.375,
|
| 945 |
+
"learning_rate": 3.3903133903133903e-07,
|
| 946 |
+
"logits/generated": 0.24424830079078674,
|
| 947 |
+
"logits/real": -0.2476576864719391,
|
| 948 |
+
"logps/generated": -556.5860595703125,
|
| 949 |
+
"logps/real": -447.5859375,
|
| 950 |
+
"loss": 0.2816,
|
| 951 |
+
"rewards/accuracies": 0.925000011920929,
|
| 952 |
+
"rewards/generated": -0.7970044612884521,
|
| 953 |
+
"rewards/margins": 0.8752428889274597,
|
| 954 |
+
"rewards/real": 0.0782383605837822,
|
| 955 |
+
"step": 305
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.4,
|
| 959 |
+
"grad_norm": 24.875,
|
| 960 |
+
"learning_rate": 3.354700854700854e-07,
|
| 961 |
+
"logits/generated": 0.25422555208206177,
|
| 962 |
+
"logits/real": -0.27572309970855713,
|
| 963 |
+
"logps/generated": -581.782958984375,
|
| 964 |
+
"logps/real": -469.9566955566406,
|
| 965 |
+
"loss": 0.2844,
|
| 966 |
+
"rewards/accuracies": 0.9375,
|
| 967 |
+
"rewards/generated": -0.8562766313552856,
|
| 968 |
+
"rewards/margins": 0.9663234949111938,
|
| 969 |
+
"rewards/real": 0.11004678159952164,
|
| 970 |
+
"step": 310
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.4,
|
| 974 |
+
"grad_norm": 23.875,
|
| 975 |
+
"learning_rate": 3.319088319088319e-07,
|
| 976 |
+
"logits/generated": 0.2173062115907669,
|
| 977 |
+
"logits/real": -0.2551296651363373,
|
| 978 |
+
"logps/generated": -550.9298706054688,
|
| 979 |
+
"logps/real": -478.9058532714844,
|
| 980 |
+
"loss": 0.3036,
|
| 981 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 982 |
+
"rewards/generated": -0.7514694333076477,
|
| 983 |
+
"rewards/margins": 0.7880412340164185,
|
| 984 |
+
"rewards/real": 0.03657188266515732,
|
| 985 |
+
"step": 315
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.41,
|
| 989 |
+
"grad_norm": 25.625,
|
| 990 |
+
"learning_rate": 3.2834757834757833e-07,
|
| 991 |
+
"logits/generated": 0.27889084815979004,
|
| 992 |
+
"logits/real": -0.19512683153152466,
|
| 993 |
+
"logps/generated": -569.74560546875,
|
| 994 |
+
"logps/real": -422.5086364746094,
|
| 995 |
+
"loss": 0.2891,
|
| 996 |
+
"rewards/accuracies": 0.949999988079071,
|
| 997 |
+
"rewards/generated": -0.8094115257263184,
|
| 998 |
+
"rewards/margins": 0.8925756216049194,
|
| 999 |
+
"rewards/real": 0.08316410332918167,
|
| 1000 |
+
"step": 320
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.42,
|
| 1004 |
+
"grad_norm": 25.125,
|
| 1005 |
+
"learning_rate": 3.2478632478632476e-07,
|
| 1006 |
+
"logits/generated": 0.2879219949245453,
|
| 1007 |
+
"logits/real": -0.27906739711761475,
|
| 1008 |
+
"logps/generated": -583.55908203125,
|
| 1009 |
+
"logps/real": -479.68988037109375,
|
| 1010 |
+
"loss": 0.2798,
|
| 1011 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1012 |
+
"rewards/generated": -0.7989848852157593,
|
| 1013 |
+
"rewards/margins": 0.9030762910842896,
|
| 1014 |
+
"rewards/real": 0.10409130901098251,
|
| 1015 |
+
"step": 325
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 0.42,
|
| 1019 |
+
"grad_norm": 23.125,
|
| 1020 |
+
"learning_rate": 3.212250712250712e-07,
|
| 1021 |
+
"logits/generated": 0.2541460394859314,
|
| 1022 |
+
"logits/real": -0.21789531409740448,
|
| 1023 |
+
"logps/generated": -570.9312744140625,
|
| 1024 |
+
"logps/real": -437.07476806640625,
|
| 1025 |
+
"loss": 0.2908,
|
| 1026 |
+
"rewards/accuracies": 0.925000011920929,
|
| 1027 |
+
"rewards/generated": -0.8303514719009399,
|
| 1028 |
+
"rewards/margins": 0.9341219663619995,
|
| 1029 |
+
"rewards/real": 0.10377053916454315,
|
| 1030 |
+
"step": 330
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.43,
|
| 1034 |
+
"grad_norm": 25.875,
|
| 1035 |
+
"learning_rate": 3.176638176638177e-07,
|
| 1036 |
+
"logits/generated": 0.2649766802787781,
|
| 1037 |
+
"logits/real": -0.23893487453460693,
|
| 1038 |
+
"logps/generated": -601.1578979492188,
|
| 1039 |
+
"logps/real": -482.6011657714844,
|
| 1040 |
+
"loss": 0.2859,
|
| 1041 |
+
"rewards/accuracies": 0.918749988079071,
|
| 1042 |
+
"rewards/generated": -0.8597179651260376,
|
| 1043 |
+
"rewards/margins": 0.926683247089386,
|
| 1044 |
+
"rewards/real": 0.06696534156799316,
|
| 1045 |
+
"step": 335
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.44,
|
| 1049 |
+
"grad_norm": 26.75,
|
| 1050 |
+
"learning_rate": 3.1410256410256407e-07,
|
| 1051 |
+
"logits/generated": 0.2579233646392822,
|
| 1052 |
+
"logits/real": -0.2822721600532532,
|
| 1053 |
+
"logps/generated": -575.4134521484375,
|
| 1054 |
+
"logps/real": -461.46099853515625,
|
| 1055 |
+
"loss": 0.2751,
|
| 1056 |
+
"rewards/accuracies": 0.956250011920929,
|
| 1057 |
+
"rewards/generated": -0.8965667486190796,
|
| 1058 |
+
"rewards/margins": 0.9989431500434875,
|
| 1059 |
+
"rewards/real": 0.10237620025873184,
|
| 1060 |
+
"step": 340
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.44,
|
| 1064 |
+
"grad_norm": 24.125,
|
| 1065 |
+
"learning_rate": 3.1054131054131055e-07,
|
| 1066 |
+
"logits/generated": 0.24282464385032654,
|
| 1067 |
+
"logits/real": -0.19455142319202423,
|
| 1068 |
+
"logps/generated": -573.3221435546875,
|
| 1069 |
+
"logps/real": -401.08319091796875,
|
| 1070 |
+
"loss": 0.2932,
|
| 1071 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1072 |
+
"rewards/generated": -0.7664304971694946,
|
| 1073 |
+
"rewards/margins": 0.8475470542907715,
|
| 1074 |
+
"rewards/real": 0.08111655712127686,
|
| 1075 |
+
"step": 345
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.45,
|
| 1079 |
+
"grad_norm": 21.875,
|
| 1080 |
+
"learning_rate": 3.06980056980057e-07,
|
| 1081 |
+
"logits/generated": 0.2672547698020935,
|
| 1082 |
+
"logits/real": -0.2481343299150467,
|
| 1083 |
+
"logps/generated": -546.6334228515625,
|
| 1084 |
+
"logps/real": -449.42694091796875,
|
| 1085 |
+
"loss": 0.2879,
|
| 1086 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1087 |
+
"rewards/generated": -0.8173602223396301,
|
| 1088 |
+
"rewards/margins": 0.8795957565307617,
|
| 1089 |
+
"rewards/real": 0.062235575169324875,
|
| 1090 |
+
"step": 350
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.45,
|
| 1094 |
+
"grad_norm": 23.25,
|
| 1095 |
+
"learning_rate": 3.0341880341880337e-07,
|
| 1096 |
+
"logits/generated": 0.2704509198665619,
|
| 1097 |
+
"logits/real": -0.26537832617759705,
|
| 1098 |
+
"logps/generated": -572.9697265625,
|
| 1099 |
+
"logps/real": -483.6675720214844,
|
| 1100 |
+
"loss": 0.2776,
|
| 1101 |
+
"rewards/accuracies": 0.918749988079071,
|
| 1102 |
+
"rewards/generated": -0.8736934661865234,
|
| 1103 |
+
"rewards/margins": 0.9758409261703491,
|
| 1104 |
+
"rewards/real": 0.10214747488498688,
|
| 1105 |
+
"step": 355
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.46,
|
| 1109 |
+
"grad_norm": 24.375,
|
| 1110 |
+
"learning_rate": 2.9985754985754986e-07,
|
| 1111 |
+
"logits/generated": 0.23115357756614685,
|
| 1112 |
+
"logits/real": -0.21211636066436768,
|
| 1113 |
+
"logps/generated": -565.6680908203125,
|
| 1114 |
+
"logps/real": -419.73004150390625,
|
| 1115 |
+
"loss": 0.2911,
|
| 1116 |
+
"rewards/accuracies": 0.918749988079071,
|
| 1117 |
+
"rewards/generated": -0.8500432968139648,
|
| 1118 |
+
"rewards/margins": 0.9019571542739868,
|
| 1119 |
+
"rewards/real": 0.05191392824053764,
|
| 1120 |
+
"step": 360
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.47,
|
| 1124 |
+
"grad_norm": 24.0,
|
| 1125 |
+
"learning_rate": 2.962962962962963e-07,
|
| 1126 |
+
"logits/generated": 0.24056752026081085,
|
| 1127 |
+
"logits/real": -0.26113012433052063,
|
| 1128 |
+
"logps/generated": -570.0296020507812,
|
| 1129 |
+
"logps/real": -483.2586975097656,
|
| 1130 |
+
"loss": 0.2824,
|
| 1131 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1132 |
+
"rewards/generated": -0.8894678354263306,
|
| 1133 |
+
"rewards/margins": 0.985159695148468,
|
| 1134 |
+
"rewards/real": 0.09569184482097626,
|
| 1135 |
+
"step": 365
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.47,
|
| 1139 |
+
"grad_norm": 24.75,
|
| 1140 |
+
"learning_rate": 2.927350427350427e-07,
|
| 1141 |
+
"logits/generated": 0.23835477232933044,
|
| 1142 |
+
"logits/real": -0.2653237283229828,
|
| 1143 |
+
"logps/generated": -572.784912109375,
|
| 1144 |
+
"logps/real": -469.38995361328125,
|
| 1145 |
+
"loss": 0.2879,
|
| 1146 |
+
"rewards/accuracies": 0.925000011920929,
|
| 1147 |
+
"rewards/generated": -0.9159420132637024,
|
| 1148 |
+
"rewards/margins": 0.9773737192153931,
|
| 1149 |
+
"rewards/real": 0.061431728303432465,
|
| 1150 |
+
"step": 370
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.48,
|
| 1154 |
+
"grad_norm": 21.25,
|
| 1155 |
+
"learning_rate": 2.8917378917378916e-07,
|
| 1156 |
+
"logits/generated": 0.23210255801677704,
|
| 1157 |
+
"logits/real": -0.18087713420391083,
|
| 1158 |
+
"logps/generated": -559.0404663085938,
|
| 1159 |
+
"logps/real": -409.7191467285156,
|
| 1160 |
+
"loss": 0.2834,
|
| 1161 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1162 |
+
"rewards/generated": -0.861811637878418,
|
| 1163 |
+
"rewards/margins": 0.9838217496871948,
|
| 1164 |
+
"rewards/real": 0.12201017141342163,
|
| 1165 |
+
"step": 375
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.49,
|
| 1169 |
+
"grad_norm": 23.375,
|
| 1170 |
+
"learning_rate": 2.8561253561253565e-07,
|
| 1171 |
+
"logits/generated": 0.21506235003471375,
|
| 1172 |
+
"logits/real": -0.26088595390319824,
|
| 1173 |
+
"logps/generated": -571.0586547851562,
|
| 1174 |
+
"logps/real": -450.5771484375,
|
| 1175 |
+
"loss": 0.2905,
|
| 1176 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1177 |
+
"rewards/generated": -0.8820127248764038,
|
| 1178 |
+
"rewards/margins": 0.9079700708389282,
|
| 1179 |
+
"rewards/real": 0.025957267731428146,
|
| 1180 |
+
"step": 380
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.49,
|
| 1184 |
+
"grad_norm": 23.0,
|
| 1185 |
+
"learning_rate": 2.8205128205128203e-07,
|
| 1186 |
+
"logits/generated": 0.22029800713062286,
|
| 1187 |
+
"logits/real": -0.25589895248413086,
|
| 1188 |
+
"logps/generated": -549.1640625,
|
| 1189 |
+
"logps/real": -407.2776794433594,
|
| 1190 |
+
"loss": 0.2792,
|
| 1191 |
+
"rewards/accuracies": 0.9125000238418579,
|
| 1192 |
+
"rewards/generated": -0.8181031942367554,
|
| 1193 |
+
"rewards/margins": 0.9209718704223633,
|
| 1194 |
+
"rewards/real": 0.1028686910867691,
|
| 1195 |
+
"step": 385
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 0.5,
|
| 1199 |
+
"grad_norm": 23.125,
|
| 1200 |
+
"learning_rate": 2.7849002849002846e-07,
|
| 1201 |
+
"logits/generated": 0.2413676679134369,
|
| 1202 |
+
"logits/real": -0.2968316078186035,
|
| 1203 |
+
"logps/generated": -567.1182861328125,
|
| 1204 |
+
"logps/real": -498.888671875,
|
| 1205 |
+
"loss": 0.2883,
|
| 1206 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1207 |
+
"rewards/generated": -0.8102291822433472,
|
| 1208 |
+
"rewards/margins": 0.8724681735038757,
|
| 1209 |
+
"rewards/real": 0.062238991260528564,
|
| 1210 |
+
"step": 390
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.51,
|
| 1214 |
+
"grad_norm": 24.375,
|
| 1215 |
+
"learning_rate": 2.749287749287749e-07,
|
| 1216 |
+
"logits/generated": 0.25530534982681274,
|
| 1217 |
+
"logits/real": -0.27374374866485596,
|
| 1218 |
+
"logps/generated": -560.980712890625,
|
| 1219 |
+
"logps/real": -457.9765625,
|
| 1220 |
+
"loss": 0.2639,
|
| 1221 |
+
"rewards/accuracies": 0.956250011920929,
|
| 1222 |
+
"rewards/generated": -0.9177068471908569,
|
| 1223 |
+
"rewards/margins": 1.0232750177383423,
|
| 1224 |
+
"rewards/real": 0.10556807368993759,
|
| 1225 |
+
"step": 395
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.51,
|
| 1229 |
+
"grad_norm": 23.5,
|
| 1230 |
+
"learning_rate": 2.7136752136752133e-07,
|
| 1231 |
+
"logits/generated": 0.2365385740995407,
|
| 1232 |
+
"logits/real": -0.2824167311191559,
|
| 1233 |
+
"logps/generated": -593.1279296875,
|
| 1234 |
+
"logps/real": -420.54608154296875,
|
| 1235 |
+
"loss": 0.2769,
|
| 1236 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1237 |
+
"rewards/generated": -0.9472834467887878,
|
| 1238 |
+
"rewards/margins": 1.0492109060287476,
|
| 1239 |
+
"rewards/real": 0.1019275039434433,
|
| 1240 |
+
"step": 400
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.52,
|
| 1244 |
+
"grad_norm": 23.5,
|
| 1245 |
+
"learning_rate": 2.678062678062678e-07,
|
| 1246 |
+
"logits/generated": 0.2417086362838745,
|
| 1247 |
+
"logits/real": -0.21366417407989502,
|
| 1248 |
+
"logps/generated": -576.4993896484375,
|
| 1249 |
+
"logps/real": -450.2381896972656,
|
| 1250 |
+
"loss": 0.2767,
|
| 1251 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1252 |
+
"rewards/generated": -0.8732357025146484,
|
| 1253 |
+
"rewards/margins": 0.9638237953186035,
|
| 1254 |
+
"rewards/real": 0.0905880555510521,
|
| 1255 |
+
"step": 405
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.52,
|
| 1259 |
+
"grad_norm": 24.375,
|
| 1260 |
+
"learning_rate": 2.642450142450142e-07,
|
| 1261 |
+
"logits/generated": 0.2748910188674927,
|
| 1262 |
+
"logits/real": -0.2791479229927063,
|
| 1263 |
+
"logps/generated": -584.0994262695312,
|
| 1264 |
+
"logps/real": -474.29632568359375,
|
| 1265 |
+
"loss": 0.2624,
|
| 1266 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1267 |
+
"rewards/generated": -0.9621374011039734,
|
| 1268 |
+
"rewards/margins": 1.0574219226837158,
|
| 1269 |
+
"rewards/real": 0.09528439491987228,
|
| 1270 |
+
"step": 410
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.53,
|
| 1274 |
+
"grad_norm": 22.0,
|
| 1275 |
+
"learning_rate": 2.606837606837607e-07,
|
| 1276 |
+
"logits/generated": 0.24150626361370087,
|
| 1277 |
+
"logits/real": -0.2777560353279114,
|
| 1278 |
+
"logps/generated": -580.0728759765625,
|
| 1279 |
+
"logps/real": -422.99273681640625,
|
| 1280 |
+
"loss": 0.27,
|
| 1281 |
+
"rewards/accuracies": 0.981249988079071,
|
| 1282 |
+
"rewards/generated": -0.893054187297821,
|
| 1283 |
+
"rewards/margins": 1.0188217163085938,
|
| 1284 |
+
"rewards/real": 0.12576748430728912,
|
| 1285 |
+
"step": 415
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.54,
|
| 1289 |
+
"grad_norm": 23.125,
|
| 1290 |
+
"learning_rate": 2.571225071225071e-07,
|
| 1291 |
+
"logits/generated": 0.2542496621608734,
|
| 1292 |
+
"logits/real": -0.19691842794418335,
|
| 1293 |
+
"logps/generated": -562.2733154296875,
|
| 1294 |
+
"logps/real": -425.752685546875,
|
| 1295 |
+
"loss": 0.2695,
|
| 1296 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1297 |
+
"rewards/generated": -0.9875919222831726,
|
| 1298 |
+
"rewards/margins": 1.0910958051681519,
|
| 1299 |
+
"rewards/real": 0.10350396484136581,
|
| 1300 |
+
"step": 420
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 0.54,
|
| 1304 |
+
"grad_norm": 23.125,
|
| 1305 |
+
"learning_rate": 2.5356125356125355e-07,
|
| 1306 |
+
"logits/generated": 0.20040163397789001,
|
| 1307 |
+
"logits/real": -0.2452303171157837,
|
| 1308 |
+
"logps/generated": -573.46142578125,
|
| 1309 |
+
"logps/real": -419.46795654296875,
|
| 1310 |
+
"loss": 0.264,
|
| 1311 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1312 |
+
"rewards/generated": -0.9656850099563599,
|
| 1313 |
+
"rewards/margins": 1.063493013381958,
|
| 1314 |
+
"rewards/real": 0.0978081002831459,
|
| 1315 |
+
"step": 425
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 0.55,
|
| 1319 |
+
"grad_norm": 24.875,
|
| 1320 |
+
"learning_rate": 2.5e-07,
|
| 1321 |
+
"logits/generated": 0.2523021697998047,
|
| 1322 |
+
"logits/real": -0.2764134407043457,
|
| 1323 |
+
"logps/generated": -564.8800048828125,
|
| 1324 |
+
"logps/real": -534.353759765625,
|
| 1325 |
+
"loss": 0.2699,
|
| 1326 |
+
"rewards/accuracies": 0.90625,
|
| 1327 |
+
"rewards/generated": -0.856059730052948,
|
| 1328 |
+
"rewards/margins": 0.9356383085250854,
|
| 1329 |
+
"rewards/real": 0.07957860082387924,
|
| 1330 |
+
"step": 430
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.56,
|
| 1334 |
+
"grad_norm": 25.75,
|
| 1335 |
+
"learning_rate": 2.464387464387464e-07,
|
| 1336 |
+
"logits/generated": 0.24086351692676544,
|
| 1337 |
+
"logits/real": -0.3009343147277832,
|
| 1338 |
+
"logps/generated": -557.8231201171875,
|
| 1339 |
+
"logps/real": -470.4691467285156,
|
| 1340 |
+
"loss": 0.2697,
|
| 1341 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1342 |
+
"rewards/generated": -0.9576956629753113,
|
| 1343 |
+
"rewards/margins": 1.0484102964401245,
|
| 1344 |
+
"rewards/real": 0.09071467071771622,
|
| 1345 |
+
"step": 435
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.56,
|
| 1349 |
+
"grad_norm": 23.375,
|
| 1350 |
+
"learning_rate": 2.4287749287749286e-07,
|
| 1351 |
+
"logits/generated": 0.25982385873794556,
|
| 1352 |
+
"logits/real": -0.27899274230003357,
|
| 1353 |
+
"logps/generated": -618.6469116210938,
|
| 1354 |
+
"logps/real": -502.1651916503906,
|
| 1355 |
+
"loss": 0.269,
|
| 1356 |
+
"rewards/accuracies": 0.925000011920929,
|
| 1357 |
+
"rewards/generated": -0.9170626401901245,
|
| 1358 |
+
"rewards/margins": 1.0099645853042603,
|
| 1359 |
+
"rewards/real": 0.09290206432342529,
|
| 1360 |
+
"step": 440
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.57,
|
| 1364 |
+
"grad_norm": 21.5,
|
| 1365 |
+
"learning_rate": 2.393162393162393e-07,
|
| 1366 |
+
"logits/generated": 0.21476399898529053,
|
| 1367 |
+
"logits/real": -0.2801568806171417,
|
| 1368 |
+
"logps/generated": -586.9801025390625,
|
| 1369 |
+
"logps/real": -428.6065979003906,
|
| 1370 |
+
"loss": 0.2744,
|
| 1371 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1372 |
+
"rewards/generated": -0.8958868980407715,
|
| 1373 |
+
"rewards/margins": 0.9335384368896484,
|
| 1374 |
+
"rewards/real": 0.03765156865119934,
|
| 1375 |
+
"step": 445
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.58,
|
| 1379 |
+
"grad_norm": 21.125,
|
| 1380 |
+
"learning_rate": 2.3575498575498575e-07,
|
| 1381 |
+
"logits/generated": 0.22820791602134705,
|
| 1382 |
+
"logits/real": -0.2373519241809845,
|
| 1383 |
+
"logps/generated": -591.01416015625,
|
| 1384 |
+
"logps/real": -455.0992736816406,
|
| 1385 |
+
"loss": 0.261,
|
| 1386 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1387 |
+
"rewards/generated": -0.9810103178024292,
|
| 1388 |
+
"rewards/margins": 1.0630879402160645,
|
| 1389 |
+
"rewards/real": 0.08207770437002182,
|
| 1390 |
+
"step": 450
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.58,
|
| 1394 |
+
"grad_norm": 23.625,
|
| 1395 |
+
"learning_rate": 2.3219373219373216e-07,
|
| 1396 |
+
"logits/generated": 0.2243388593196869,
|
| 1397 |
+
"logits/real": -0.3200889527797699,
|
| 1398 |
+
"logps/generated": -562.8119506835938,
|
| 1399 |
+
"logps/real": -481.96697998046875,
|
| 1400 |
+
"loss": 0.2646,
|
| 1401 |
+
"rewards/accuracies": 0.9375,
|
| 1402 |
+
"rewards/generated": -0.9419108629226685,
|
| 1403 |
+
"rewards/margins": 1.0310115814208984,
|
| 1404 |
+
"rewards/real": 0.08910069614648819,
|
| 1405 |
+
"step": 455
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 0.59,
|
| 1409 |
+
"grad_norm": 22.625,
|
| 1410 |
+
"learning_rate": 2.2863247863247862e-07,
|
| 1411 |
+
"logits/generated": 0.24365122616291046,
|
| 1412 |
+
"logits/real": -0.2204679548740387,
|
| 1413 |
+
"logps/generated": -557.0007934570312,
|
| 1414 |
+
"logps/real": -446.1201171875,
|
| 1415 |
+
"loss": 0.2557,
|
| 1416 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1417 |
+
"rewards/generated": -0.9837862849235535,
|
| 1418 |
+
"rewards/margins": 1.0443240404129028,
|
| 1419 |
+
"rewards/real": 0.060537658631801605,
|
| 1420 |
+
"step": 460
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.6,
|
| 1424 |
+
"grad_norm": 24.5,
|
| 1425 |
+
"learning_rate": 2.2507122507122505e-07,
|
| 1426 |
+
"logits/generated": 0.21367163956165314,
|
| 1427 |
+
"logits/real": -0.26951321959495544,
|
| 1428 |
+
"logps/generated": -555.830078125,
|
| 1429 |
+
"logps/real": -465.67864990234375,
|
| 1430 |
+
"loss": 0.2615,
|
| 1431 |
+
"rewards/accuracies": 0.925000011920929,
|
| 1432 |
+
"rewards/generated": -1.0059828758239746,
|
| 1433 |
+
"rewards/margins": 1.0972858667373657,
|
| 1434 |
+
"rewards/real": 0.09130293130874634,
|
| 1435 |
+
"step": 465
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 0.6,
|
| 1439 |
+
"grad_norm": 23.125,
|
| 1440 |
+
"learning_rate": 2.215099715099715e-07,
|
| 1441 |
+
"logits/generated": 0.22326333820819855,
|
| 1442 |
+
"logits/real": -0.266071081161499,
|
| 1443 |
+
"logps/generated": -554.1658325195312,
|
| 1444 |
+
"logps/real": -408.9630432128906,
|
| 1445 |
+
"loss": 0.2685,
|
| 1446 |
+
"rewards/accuracies": 0.9375,
|
| 1447 |
+
"rewards/generated": -0.8925862312316895,
|
| 1448 |
+
"rewards/margins": 0.9420120120048523,
|
| 1449 |
+
"rewards/real": 0.04942578822374344,
|
| 1450 |
+
"step": 470
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.61,
|
| 1454 |
+
"grad_norm": 23.25,
|
| 1455 |
+
"learning_rate": 2.1794871794871795e-07,
|
| 1456 |
+
"logits/generated": 0.24654094874858856,
|
| 1457 |
+
"logits/real": -0.2880334258079529,
|
| 1458 |
+
"logps/generated": -571.1951904296875,
|
| 1459 |
+
"logps/real": -466.7237854003906,
|
| 1460 |
+
"loss": 0.2653,
|
| 1461 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1462 |
+
"rewards/generated": -0.9611663818359375,
|
| 1463 |
+
"rewards/margins": 1.0481185913085938,
|
| 1464 |
+
"rewards/real": 0.08695221692323685,
|
| 1465 |
+
"step": 475
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.61,
|
| 1469 |
+
"grad_norm": 21.375,
|
| 1470 |
+
"learning_rate": 2.1438746438746438e-07,
|
| 1471 |
+
"logits/generated": 0.2374725341796875,
|
| 1472 |
+
"logits/real": -0.2841905653476715,
|
| 1473 |
+
"logps/generated": -597.453125,
|
| 1474 |
+
"logps/real": -467.695068359375,
|
| 1475 |
+
"loss": 0.2606,
|
| 1476 |
+
"rewards/accuracies": 0.956250011920929,
|
| 1477 |
+
"rewards/generated": -1.0608772039413452,
|
| 1478 |
+
"rewards/margins": 1.1110285520553589,
|
| 1479 |
+
"rewards/real": 0.05015145614743233,
|
| 1480 |
+
"step": 480
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.62,
|
| 1484 |
+
"grad_norm": 22.625,
|
| 1485 |
+
"learning_rate": 2.1082621082621082e-07,
|
| 1486 |
+
"logits/generated": 0.20874783396720886,
|
| 1487 |
+
"logits/real": -0.24746723473072052,
|
| 1488 |
+
"logps/generated": -558.3968505859375,
|
| 1489 |
+
"logps/real": -418.06085205078125,
|
| 1490 |
+
"loss": 0.2779,
|
| 1491 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1492 |
+
"rewards/generated": -0.9692691564559937,
|
| 1493 |
+
"rewards/margins": 1.016911268234253,
|
| 1494 |
+
"rewards/real": 0.04764218255877495,
|
| 1495 |
+
"step": 485
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.63,
|
| 1499 |
+
"grad_norm": 21.5,
|
| 1500 |
+
"learning_rate": 2.0726495726495728e-07,
|
| 1501 |
+
"logits/generated": 0.23431365191936493,
|
| 1502 |
+
"logits/real": -0.3022805154323578,
|
| 1503 |
+
"logps/generated": -560.6641845703125,
|
| 1504 |
+
"logps/real": -508.0726013183594,
|
| 1505 |
+
"loss": 0.2748,
|
| 1506 |
+
"rewards/accuracies": 0.918749988079071,
|
| 1507 |
+
"rewards/generated": -0.9711412191390991,
|
| 1508 |
+
"rewards/margins": 0.9574298858642578,
|
| 1509 |
+
"rewards/real": -0.013711368665099144,
|
| 1510 |
+
"step": 490
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 0.63,
|
| 1514 |
+
"grad_norm": 21.875,
|
| 1515 |
+
"learning_rate": 2.0370370370370369e-07,
|
| 1516 |
+
"logits/generated": 0.217031329870224,
|
| 1517 |
+
"logits/real": -0.2614264488220215,
|
| 1518 |
+
"logps/generated": -559.6924438476562,
|
| 1519 |
+
"logps/real": -446.6640625,
|
| 1520 |
+
"loss": 0.2687,
|
| 1521 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1522 |
+
"rewards/generated": -0.9257688522338867,
|
| 1523 |
+
"rewards/margins": 0.9959582090377808,
|
| 1524 |
+
"rewards/real": 0.07018940150737762,
|
| 1525 |
+
"step": 495
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 0.64,
|
| 1529 |
+
"grad_norm": 22.625,
|
| 1530 |
+
"learning_rate": 2.0014245014245012e-07,
|
| 1531 |
+
"logits/generated": 0.24876682460308075,
|
| 1532 |
+
"logits/real": -0.28565463423728943,
|
| 1533 |
+
"logps/generated": -548.2100830078125,
|
| 1534 |
+
"logps/real": -472.73223876953125,
|
| 1535 |
+
"loss": 0.2661,
|
| 1536 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1537 |
+
"rewards/generated": -0.9158692359924316,
|
| 1538 |
+
"rewards/margins": 1.0267049074172974,
|
| 1539 |
+
"rewards/real": 0.11083561182022095,
|
| 1540 |
+
"step": 500
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.65,
|
| 1544 |
+
"grad_norm": 22.75,
|
| 1545 |
+
"learning_rate": 1.9658119658119656e-07,
|
| 1546 |
+
"logits/generated": 0.21912440657615662,
|
| 1547 |
+
"logits/real": -0.27414244413375854,
|
| 1548 |
+
"logps/generated": -540.6039428710938,
|
| 1549 |
+
"logps/real": -458.55975341796875,
|
| 1550 |
+
"loss": 0.2633,
|
| 1551 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1552 |
+
"rewards/generated": -0.9050576090812683,
|
| 1553 |
+
"rewards/margins": 1.0112102031707764,
|
| 1554 |
+
"rewards/real": 0.10615254938602448,
|
| 1555 |
+
"step": 505
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 0.65,
|
| 1559 |
+
"grad_norm": 23.375,
|
| 1560 |
+
"learning_rate": 1.9301994301994302e-07,
|
| 1561 |
+
"logits/generated": 0.22622957825660706,
|
| 1562 |
+
"logits/real": -0.25896841287612915,
|
| 1563 |
+
"logps/generated": -567.3508911132812,
|
| 1564 |
+
"logps/real": -443.44915771484375,
|
| 1565 |
+
"loss": 0.263,
|
| 1566 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1567 |
+
"rewards/generated": -0.9187369346618652,
|
| 1568 |
+
"rewards/margins": 1.0117876529693604,
|
| 1569 |
+
"rewards/real": 0.09305067360401154,
|
| 1570 |
+
"step": 510
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.66,
|
| 1574 |
+
"grad_norm": 24.125,
|
| 1575 |
+
"learning_rate": 1.8945868945868945e-07,
|
| 1576 |
+
"logits/generated": 0.22547177970409393,
|
| 1577 |
+
"logits/real": -0.30368226766586304,
|
| 1578 |
+
"logps/generated": -568.8248291015625,
|
| 1579 |
+
"logps/real": -470.2293395996094,
|
| 1580 |
+
"loss": 0.2686,
|
| 1581 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1582 |
+
"rewards/generated": -1.009271502494812,
|
| 1583 |
+
"rewards/margins": 1.0391043424606323,
|
| 1584 |
+
"rewards/real": 0.029832839965820312,
|
| 1585 |
+
"step": 515
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.67,
|
| 1589 |
+
"grad_norm": 23.625,
|
| 1590 |
+
"learning_rate": 1.8589743589743588e-07,
|
| 1591 |
+
"logits/generated": 0.228875070810318,
|
| 1592 |
+
"logits/real": -0.27519601583480835,
|
| 1593 |
+
"logps/generated": -548.3033447265625,
|
| 1594 |
+
"logps/real": -456.94049072265625,
|
| 1595 |
+
"loss": 0.2579,
|
| 1596 |
+
"rewards/accuracies": 0.90625,
|
| 1597 |
+
"rewards/generated": -0.9734387397766113,
|
| 1598 |
+
"rewards/margins": 1.06186842918396,
|
| 1599 |
+
"rewards/real": 0.08842964470386505,
|
| 1600 |
+
"step": 520
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.67,
|
| 1604 |
+
"grad_norm": 22.25,
|
| 1605 |
+
"learning_rate": 1.8233618233618234e-07,
|
| 1606 |
+
"logits/generated": 0.24410124123096466,
|
| 1607 |
+
"logits/real": -0.24580614268779755,
|
| 1608 |
+
"logps/generated": -526.7844848632812,
|
| 1609 |
+
"logps/real": -461.39599609375,
|
| 1610 |
+
"loss": 0.2555,
|
| 1611 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1612 |
+
"rewards/generated": -0.9268838763237,
|
| 1613 |
+
"rewards/margins": 1.0250850915908813,
|
| 1614 |
+
"rewards/real": 0.09820131957530975,
|
| 1615 |
+
"step": 525
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 0.68,
|
| 1619 |
+
"grad_norm": 22.75,
|
| 1620 |
+
"learning_rate": 1.7877492877492878e-07,
|
| 1621 |
+
"logits/generated": 0.24622035026550293,
|
| 1622 |
+
"logits/real": -0.350544810295105,
|
| 1623 |
+
"logps/generated": -583.2838134765625,
|
| 1624 |
+
"logps/real": -484.83758544921875,
|
| 1625 |
+
"loss": 0.2696,
|
| 1626 |
+
"rewards/accuracies": 0.9375,
|
| 1627 |
+
"rewards/generated": -1.0305290222167969,
|
| 1628 |
+
"rewards/margins": 1.0726959705352783,
|
| 1629 |
+
"rewards/real": 0.04216689243912697,
|
| 1630 |
+
"step": 530
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 0.68,
|
| 1634 |
+
"grad_norm": 20.625,
|
| 1635 |
+
"learning_rate": 1.752136752136752e-07,
|
| 1636 |
+
"logits/generated": 0.23964126408100128,
|
| 1637 |
+
"logits/real": -0.21912190318107605,
|
| 1638 |
+
"logps/generated": -574.9818115234375,
|
| 1639 |
+
"logps/real": -413.7064514160156,
|
| 1640 |
+
"loss": 0.2533,
|
| 1641 |
+
"rewards/accuracies": 0.96875,
|
| 1642 |
+
"rewards/generated": -1.0674055814743042,
|
| 1643 |
+
"rewards/margins": 1.1347934007644653,
|
| 1644 |
+
"rewards/real": 0.06738778948783875,
|
| 1645 |
+
"step": 535
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 0.69,
|
| 1649 |
+
"grad_norm": 21.375,
|
| 1650 |
+
"learning_rate": 1.7165242165242165e-07,
|
| 1651 |
+
"logits/generated": 0.2101718634366989,
|
| 1652 |
+
"logits/real": -0.2580256760120392,
|
| 1653 |
+
"logps/generated": -591.0747680664062,
|
| 1654 |
+
"logps/real": -402.9712219238281,
|
| 1655 |
+
"loss": 0.2586,
|
| 1656 |
+
"rewards/accuracies": 0.90625,
|
| 1657 |
+
"rewards/generated": -0.9934548139572144,
|
| 1658 |
+
"rewards/margins": 1.0756797790527344,
|
| 1659 |
+
"rewards/real": 0.08222486823797226,
|
| 1660 |
+
"step": 540
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 0.7,
|
| 1664 |
+
"grad_norm": 22.75,
|
| 1665 |
+
"learning_rate": 1.6809116809116808e-07,
|
| 1666 |
+
"logits/generated": 0.1858968287706375,
|
| 1667 |
+
"logits/real": -0.28893381357192993,
|
| 1668 |
+
"logps/generated": -551.6287841796875,
|
| 1669 |
+
"logps/real": -424.68243408203125,
|
| 1670 |
+
"loss": 0.2614,
|
| 1671 |
+
"rewards/accuracies": 0.9125000238418579,
|
| 1672 |
+
"rewards/generated": -0.9353107213973999,
|
| 1673 |
+
"rewards/margins": 1.0308815240859985,
|
| 1674 |
+
"rewards/real": 0.09557078033685684,
|
| 1675 |
+
"step": 545
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7,
|
| 1679 |
+
"grad_norm": 22.0,
|
| 1680 |
+
"learning_rate": 1.6452991452991452e-07,
|
| 1681 |
+
"logits/generated": 0.23100645840168,
|
| 1682 |
+
"logits/real": -0.2850993275642395,
|
| 1683 |
+
"logps/generated": -597.5736083984375,
|
| 1684 |
+
"logps/real": -406.21990966796875,
|
| 1685 |
+
"loss": 0.2458,
|
| 1686 |
+
"rewards/accuracies": 0.96875,
|
| 1687 |
+
"rewards/generated": -1.1093555688858032,
|
| 1688 |
+
"rewards/margins": 1.2105884552001953,
|
| 1689 |
+
"rewards/real": 0.1012328714132309,
|
| 1690 |
+
"step": 550
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.71,
|
| 1694 |
+
"grad_norm": 22.375,
|
| 1695 |
+
"learning_rate": 1.6096866096866095e-07,
|
| 1696 |
+
"logits/generated": 0.22843961417675018,
|
| 1697 |
+
"logits/real": -0.2411879003047943,
|
| 1698 |
+
"logps/generated": -565.7466430664062,
|
| 1699 |
+
"logps/real": -450.0619201660156,
|
| 1700 |
+
"loss": 0.2582,
|
| 1701 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1702 |
+
"rewards/generated": -1.010749340057373,
|
| 1703 |
+
"rewards/margins": 1.1109167337417603,
|
| 1704 |
+
"rewards/real": 0.10016737133264542,
|
| 1705 |
+
"step": 555
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.72,
|
| 1709 |
+
"grad_norm": 22.25,
|
| 1710 |
+
"learning_rate": 1.574074074074074e-07,
|
| 1711 |
+
"logits/generated": 0.24150070548057556,
|
| 1712 |
+
"logits/real": -0.21642279624938965,
|
| 1713 |
+
"logps/generated": -569.0734252929688,
|
| 1714 |
+
"logps/real": -448.4789123535156,
|
| 1715 |
+
"loss": 0.2589,
|
| 1716 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1717 |
+
"rewards/generated": -1.0423076152801514,
|
| 1718 |
+
"rewards/margins": 1.118851900100708,
|
| 1719 |
+
"rewards/real": 0.07654404640197754,
|
| 1720 |
+
"step": 560
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 0.72,
|
| 1724 |
+
"grad_norm": 23.625,
|
| 1725 |
+
"learning_rate": 1.5384615384615385e-07,
|
| 1726 |
+
"logits/generated": 0.18384118378162384,
|
| 1727 |
+
"logits/real": -0.2612574100494385,
|
| 1728 |
+
"logps/generated": -577.3256225585938,
|
| 1729 |
+
"logps/real": -445.0957946777344,
|
| 1730 |
+
"loss": 0.2617,
|
| 1731 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1732 |
+
"rewards/generated": -1.027761697769165,
|
| 1733 |
+
"rewards/margins": 1.0843464136123657,
|
| 1734 |
+
"rewards/real": 0.056584786623716354,
|
| 1735 |
+
"step": 565
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 0.73,
|
| 1739 |
+
"grad_norm": 21.625,
|
| 1740 |
+
"learning_rate": 1.5028490028490028e-07,
|
| 1741 |
+
"logits/generated": 0.21284691989421844,
|
| 1742 |
+
"logits/real": -0.27072662115097046,
|
| 1743 |
+
"logps/generated": -576.1101684570312,
|
| 1744 |
+
"logps/real": -405.94073486328125,
|
| 1745 |
+
"loss": 0.2469,
|
| 1746 |
+
"rewards/accuracies": 0.956250011920929,
|
| 1747 |
+
"rewards/generated": -0.9797765612602234,
|
| 1748 |
+
"rewards/margins": 1.1207729578018188,
|
| 1749 |
+
"rewards/real": 0.14099633693695068,
|
| 1750 |
+
"step": 570
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 0.74,
|
| 1754 |
+
"grad_norm": 18.625,
|
| 1755 |
+
"learning_rate": 1.4672364672364671e-07,
|
| 1756 |
+
"logits/generated": 0.2502863109111786,
|
| 1757 |
+
"logits/real": -0.2541927099227905,
|
| 1758 |
+
"logps/generated": -583.0015869140625,
|
| 1759 |
+
"logps/real": -480.0306701660156,
|
| 1760 |
+
"loss": 0.2581,
|
| 1761 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1762 |
+
"rewards/generated": -0.9324277639389038,
|
| 1763 |
+
"rewards/margins": 1.014402151107788,
|
| 1764 |
+
"rewards/real": 0.08197424560785294,
|
| 1765 |
+
"step": 575
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 0.74,
|
| 1769 |
+
"grad_norm": 22.25,
|
| 1770 |
+
"learning_rate": 1.4316239316239315e-07,
|
| 1771 |
+
"logits/generated": 0.17801856994628906,
|
| 1772 |
+
"logits/real": -0.29988187551498413,
|
| 1773 |
+
"logps/generated": -559.722412109375,
|
| 1774 |
+
"logps/real": -396.5543518066406,
|
| 1775 |
+
"loss": 0.2549,
|
| 1776 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1777 |
+
"rewards/generated": -1.0840636491775513,
|
| 1778 |
+
"rewards/margins": 1.1837726831436157,
|
| 1779 |
+
"rewards/real": 0.09970910847187042,
|
| 1780 |
+
"step": 580
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.75,
|
| 1784 |
+
"grad_norm": 23.375,
|
| 1785 |
+
"learning_rate": 1.3960113960113958e-07,
|
| 1786 |
+
"logits/generated": 0.19342169165611267,
|
| 1787 |
+
"logits/real": -0.30865079164505005,
|
| 1788 |
+
"logps/generated": -531.6026611328125,
|
| 1789 |
+
"logps/real": -461.54766845703125,
|
| 1790 |
+
"loss": 0.2586,
|
| 1791 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1792 |
+
"rewards/generated": -0.9994919896125793,
|
| 1793 |
+
"rewards/margins": 1.0923388004302979,
|
| 1794 |
+
"rewards/real": 0.09284672886133194,
|
| 1795 |
+
"step": 585
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.76,
|
| 1799 |
+
"grad_norm": 24.5,
|
| 1800 |
+
"learning_rate": 1.3603988603988604e-07,
|
| 1801 |
+
"logits/generated": 0.20833972096443176,
|
| 1802 |
+
"logits/real": -0.302274614572525,
|
| 1803 |
+
"logps/generated": -581.9873046875,
|
| 1804 |
+
"logps/real": -479.486572265625,
|
| 1805 |
+
"loss": 0.2588,
|
| 1806 |
+
"rewards/accuracies": 0.981249988079071,
|
| 1807 |
+
"rewards/generated": -1.043798565864563,
|
| 1808 |
+
"rewards/margins": 1.113098382949829,
|
| 1809 |
+
"rewards/real": 0.06929990649223328,
|
| 1810 |
+
"step": 590
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.76,
|
| 1814 |
+
"grad_norm": 22.25,
|
| 1815 |
+
"learning_rate": 1.3247863247863248e-07,
|
| 1816 |
+
"logits/generated": 0.18650811910629272,
|
| 1817 |
+
"logits/real": -0.2915360927581787,
|
| 1818 |
+
"logps/generated": -559.4674682617188,
|
| 1819 |
+
"logps/real": -441.85272216796875,
|
| 1820 |
+
"loss": 0.2401,
|
| 1821 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1822 |
+
"rewards/generated": -1.024379849433899,
|
| 1823 |
+
"rewards/margins": 1.136856198310852,
|
| 1824 |
+
"rewards/real": 0.11247648298740387,
|
| 1825 |
+
"step": 595
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 0.77,
|
| 1829 |
+
"grad_norm": 20.875,
|
| 1830 |
+
"learning_rate": 1.289173789173789e-07,
|
| 1831 |
+
"logits/generated": 0.2449146956205368,
|
| 1832 |
+
"logits/real": -0.25022393465042114,
|
| 1833 |
+
"logps/generated": -576.7535400390625,
|
| 1834 |
+
"logps/real": -425.8819885253906,
|
| 1835 |
+
"loss": 0.2689,
|
| 1836 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 1837 |
+
"rewards/generated": -0.9874047040939331,
|
| 1838 |
+
"rewards/margins": 1.0526171922683716,
|
| 1839 |
+
"rewards/real": 0.06521251052618027,
|
| 1840 |
+
"step": 600
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 0.77,
|
| 1844 |
+
"eval_logits/generated": 0.22532238066196442,
|
| 1845 |
+
"eval_logits/real": -0.2819841802120209,
|
| 1846 |
+
"eval_logps/generated": -570.8114013671875,
|
| 1847 |
+
"eval_logps/real": -468.16961669921875,
|
| 1848 |
+
"eval_loss": 0.2591499090194702,
|
| 1849 |
+
"eval_rewards/accuracies": 0.9459999799728394,
|
| 1850 |
+
"eval_rewards/generated": -1.026713252067566,
|
| 1851 |
+
"eval_rewards/margins": 1.0870678424835205,
|
| 1852 |
+
"eval_rewards/real": 0.06035461276769638,
|
| 1853 |
+
"eval_runtime": 312.5704,
|
| 1854 |
+
"eval_samples_per_second": 6.399,
|
| 1855 |
+
"eval_steps_per_second": 0.8,
|
| 1856 |
+
"step": 600
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.77,
|
| 1860 |
+
"grad_norm": 20.125,
|
| 1861 |
+
"learning_rate": 1.2535612535612535e-07,
|
| 1862 |
+
"logits/generated": 0.26240265369415283,
|
| 1863 |
+
"logits/real": -0.27648916840553284,
|
| 1864 |
+
"logps/generated": -587.2186279296875,
|
| 1865 |
+
"logps/real": -469.5171813964844,
|
| 1866 |
+
"loss": 0.2429,
|
| 1867 |
+
"rewards/accuracies": 0.9375,
|
| 1868 |
+
"rewards/generated": -1.1691229343414307,
|
| 1869 |
+
"rewards/margins": 1.2391382455825806,
|
| 1870 |
+
"rewards/real": 0.07001528143882751,
|
| 1871 |
+
"step": 605
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.78,
|
| 1875 |
+
"grad_norm": 20.625,
|
| 1876 |
+
"learning_rate": 1.2179487179487178e-07,
|
| 1877 |
+
"logits/generated": 0.22506709396839142,
|
| 1878 |
+
"logits/real": -0.24815583229064941,
|
| 1879 |
+
"logps/generated": -573.8429565429688,
|
| 1880 |
+
"logps/real": -422.6187438964844,
|
| 1881 |
+
"loss": 0.2596,
|
| 1882 |
+
"rewards/accuracies": 0.96875,
|
| 1883 |
+
"rewards/generated": -1.049294352531433,
|
| 1884 |
+
"rewards/margins": 1.1830122470855713,
|
| 1885 |
+
"rewards/real": 0.13371792435646057,
|
| 1886 |
+
"step": 610
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.79,
|
| 1890 |
+
"grad_norm": 20.75,
|
| 1891 |
+
"learning_rate": 1.1823361823361823e-07,
|
| 1892 |
+
"logits/generated": 0.2133275717496872,
|
| 1893 |
+
"logits/real": -0.35235756635665894,
|
| 1894 |
+
"logps/generated": -569.0250854492188,
|
| 1895 |
+
"logps/real": -522.1539306640625,
|
| 1896 |
+
"loss": 0.2442,
|
| 1897 |
+
"rewards/accuracies": 0.981249988079071,
|
| 1898 |
+
"rewards/generated": -1.0183016061782837,
|
| 1899 |
+
"rewards/margins": 1.1286038160324097,
|
| 1900 |
+
"rewards/real": 0.11030229181051254,
|
| 1901 |
+
"step": 615
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.79,
|
| 1905 |
+
"grad_norm": 22.25,
|
| 1906 |
+
"learning_rate": 1.1467236467236467e-07,
|
| 1907 |
+
"logits/generated": 0.21725332736968994,
|
| 1908 |
+
"logits/real": -0.31801334023475647,
|
| 1909 |
+
"logps/generated": -525.4303588867188,
|
| 1910 |
+
"logps/real": -458.35955810546875,
|
| 1911 |
+
"loss": 0.2522,
|
| 1912 |
+
"rewards/accuracies": 0.96875,
|
| 1913 |
+
"rewards/generated": -1.0983850955963135,
|
| 1914 |
+
"rewards/margins": 1.1680142879486084,
|
| 1915 |
+
"rewards/real": 0.06962917000055313,
|
| 1916 |
+
"step": 620
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 0.8,
|
| 1920 |
+
"grad_norm": 22.625,
|
| 1921 |
+
"learning_rate": 1.111111111111111e-07,
|
| 1922 |
+
"logits/generated": 0.19476717710494995,
|
| 1923 |
+
"logits/real": -0.2784198224544525,
|
| 1924 |
+
"logps/generated": -539.5494384765625,
|
| 1925 |
+
"logps/real": -444.2561950683594,
|
| 1926 |
+
"loss": 0.2603,
|
| 1927 |
+
"rewards/accuracies": 0.925000011920929,
|
| 1928 |
+
"rewards/generated": -1.0229387283325195,
|
| 1929 |
+
"rewards/margins": 1.1077196598052979,
|
| 1930 |
+
"rewards/real": 0.08478085696697235,
|
| 1931 |
+
"step": 625
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 0.81,
|
| 1935 |
+
"grad_norm": 21.875,
|
| 1936 |
+
"learning_rate": 1.0754985754985754e-07,
|
| 1937 |
+
"logits/generated": 0.1993536651134491,
|
| 1938 |
+
"logits/real": -0.27307894825935364,
|
| 1939 |
+
"logps/generated": -559.6204833984375,
|
| 1940 |
+
"logps/real": -425.73779296875,
|
| 1941 |
+
"loss": 0.2509,
|
| 1942 |
+
"rewards/accuracies": 0.9375,
|
| 1943 |
+
"rewards/generated": -1.0144376754760742,
|
| 1944 |
+
"rewards/margins": 1.1288973093032837,
|
| 1945 |
+
"rewards/real": 0.11445963382720947,
|
| 1946 |
+
"step": 630
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 0.81,
|
| 1950 |
+
"grad_norm": 21.375,
|
| 1951 |
+
"learning_rate": 1.0398860398860399e-07,
|
| 1952 |
+
"logits/generated": 0.21056529879570007,
|
| 1953 |
+
"logits/real": -0.24936561286449432,
|
| 1954 |
+
"logps/generated": -586.7345581054688,
|
| 1955 |
+
"logps/real": -403.81884765625,
|
| 1956 |
+
"loss": 0.247,
|
| 1957 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1958 |
+
"rewards/generated": -1.08762526512146,
|
| 1959 |
+
"rewards/margins": 1.1714038848876953,
|
| 1960 |
+
"rewards/real": 0.08377879858016968,
|
| 1961 |
+
"step": 635
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.82,
|
| 1965 |
+
"grad_norm": 20.125,
|
| 1966 |
+
"learning_rate": 1.0042735042735042e-07,
|
| 1967 |
+
"logits/generated": 0.22550058364868164,
|
| 1968 |
+
"logits/real": -0.20315103232860565,
|
| 1969 |
+
"logps/generated": -574.2731323242188,
|
| 1970 |
+
"logps/real": -442.43280029296875,
|
| 1971 |
+
"loss": 0.2638,
|
| 1972 |
+
"rewards/accuracies": 0.9125000238418579,
|
| 1973 |
+
"rewards/generated": -1.087454080581665,
|
| 1974 |
+
"rewards/margins": 1.1403470039367676,
|
| 1975 |
+
"rewards/real": 0.05289285629987717,
|
| 1976 |
+
"step": 640
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.83,
|
| 1980 |
+
"grad_norm": 21.375,
|
| 1981 |
+
"learning_rate": 9.686609686609686e-08,
|
| 1982 |
+
"logits/generated": 0.2513348460197449,
|
| 1983 |
+
"logits/real": -0.24446964263916016,
|
| 1984 |
+
"logps/generated": -574.2137451171875,
|
| 1985 |
+
"logps/real": -461.9295959472656,
|
| 1986 |
+
"loss": 0.2621,
|
| 1987 |
+
"rewards/accuracies": 0.918749988079071,
|
| 1988 |
+
"rewards/generated": -0.9981436729431152,
|
| 1989 |
+
"rewards/margins": 1.0288463830947876,
|
| 1990 |
+
"rewards/real": 0.03070276975631714,
|
| 1991 |
+
"step": 645
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.83,
|
| 1995 |
+
"grad_norm": 21.25,
|
| 1996 |
+
"learning_rate": 9.33048433048433e-08,
|
| 1997 |
+
"logits/generated": 0.26496636867523193,
|
| 1998 |
+
"logits/real": -0.2916685938835144,
|
| 1999 |
+
"logps/generated": -551.1077880859375,
|
| 2000 |
+
"logps/real": -462.594482421875,
|
| 2001 |
+
"loss": 0.2516,
|
| 2002 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 2003 |
+
"rewards/generated": -1.0148528814315796,
|
| 2004 |
+
"rewards/margins": 1.0718117952346802,
|
| 2005 |
+
"rewards/real": 0.05695882439613342,
|
| 2006 |
+
"step": 650
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.84,
|
| 2010 |
+
"grad_norm": 23.125,
|
| 2011 |
+
"learning_rate": 8.974358974358974e-08,
|
| 2012 |
+
"logits/generated": 0.2397506982088089,
|
| 2013 |
+
"logits/real": -0.2423442304134369,
|
| 2014 |
+
"logps/generated": -585.54248046875,
|
| 2015 |
+
"logps/real": -456.861328125,
|
| 2016 |
+
"loss": 0.2585,
|
| 2017 |
+
"rewards/accuracies": 0.949999988079071,
|
| 2018 |
+
"rewards/generated": -1.0295922756195068,
|
| 2019 |
+
"rewards/margins": 1.0833699703216553,
|
| 2020 |
+
"rewards/real": 0.05377759784460068,
|
| 2021 |
+
"step": 655
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 0.84,
|
| 2025 |
+
"grad_norm": 20.625,
|
| 2026 |
+
"learning_rate": 8.618233618233619e-08,
|
| 2027 |
+
"logits/generated": 0.22315570712089539,
|
| 2028 |
+
"logits/real": -0.32908838987350464,
|
| 2029 |
+
"logps/generated": -574.5935668945312,
|
| 2030 |
+
"logps/real": -443.726806640625,
|
| 2031 |
+
"loss": 0.2626,
|
| 2032 |
+
"rewards/accuracies": 0.956250011920929,
|
| 2033 |
+
"rewards/generated": -1.041495680809021,
|
| 2034 |
+
"rewards/margins": 1.098388433456421,
|
| 2035 |
+
"rewards/real": 0.056892819702625275,
|
| 2036 |
+
"step": 660
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 0.85,
|
| 2040 |
+
"grad_norm": 22.375,
|
| 2041 |
+
"learning_rate": 8.262108262108261e-08,
|
| 2042 |
+
"logits/generated": 0.22458979487419128,
|
| 2043 |
+
"logits/real": -0.26257410645484924,
|
| 2044 |
+
"logps/generated": -587.2811279296875,
|
| 2045 |
+
"logps/real": -409.606201171875,
|
| 2046 |
+
"loss": 0.2487,
|
| 2047 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 2048 |
+
"rewards/generated": -1.0680924654006958,
|
| 2049 |
+
"rewards/margins": 1.1297204494476318,
|
| 2050 |
+
"rewards/real": 0.06162794679403305,
|
| 2051 |
+
"step": 665
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 0.86,
|
| 2055 |
+
"grad_norm": 22.0,
|
| 2056 |
+
"learning_rate": 7.905982905982906e-08,
|
| 2057 |
+
"logits/generated": 0.2559250593185425,
|
| 2058 |
+
"logits/real": -0.26228904724121094,
|
| 2059 |
+
"logps/generated": -611.9027099609375,
|
| 2060 |
+
"logps/real": -435.93634033203125,
|
| 2061 |
+
"loss": 0.2567,
|
| 2062 |
+
"rewards/accuracies": 0.981249988079071,
|
| 2063 |
+
"rewards/generated": -1.0482969284057617,
|
| 2064 |
+
"rewards/margins": 1.1413335800170898,
|
| 2065 |
+
"rewards/real": 0.09303668141365051,
|
| 2066 |
+
"step": 670
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 0.86,
|
| 2070 |
+
"grad_norm": 22.0,
|
| 2071 |
+
"learning_rate": 7.549857549857549e-08,
|
| 2072 |
+
"logits/generated": 0.21574994921684265,
|
| 2073 |
+
"logits/real": -0.3184639811515808,
|
| 2074 |
+
"logps/generated": -589.6176147460938,
|
| 2075 |
+
"logps/real": -459.68212890625,
|
| 2076 |
+
"loss": 0.2653,
|
| 2077 |
+
"rewards/accuracies": 0.949999988079071,
|
| 2078 |
+
"rewards/generated": -1.039894700050354,
|
| 2079 |
+
"rewards/margins": 1.0526447296142578,
|
| 2080 |
+
"rewards/real": 0.012749890796840191,
|
| 2081 |
+
"step": 675
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.87,
|
| 2085 |
+
"grad_norm": 19.75,
|
| 2086 |
+
"learning_rate": 7.193732193732194e-08,
|
| 2087 |
+
"logits/generated": 0.2510204613208771,
|
| 2088 |
+
"logits/real": -0.28606855869293213,
|
| 2089 |
+
"logps/generated": -576.8764038085938,
|
| 2090 |
+
"logps/real": -475.23370361328125,
|
| 2091 |
+
"loss": 0.2603,
|
| 2092 |
+
"rewards/accuracies": 0.925000011920929,
|
| 2093 |
+
"rewards/generated": -1.018733263015747,
|
| 2094 |
+
"rewards/margins": 1.0629549026489258,
|
| 2095 |
+
"rewards/real": 0.044221747666597366,
|
| 2096 |
+
"step": 680
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.88,
|
| 2100 |
+
"grad_norm": 22.125,
|
| 2101 |
+
"learning_rate": 6.837606837606839e-08,
|
| 2102 |
+
"logits/generated": 0.20441779494285583,
|
| 2103 |
+
"logits/real": -0.3156498372554779,
|
| 2104 |
+
"logps/generated": -552.9418334960938,
|
| 2105 |
+
"logps/real": -434.1715393066406,
|
| 2106 |
+
"loss": 0.2664,
|
| 2107 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 2108 |
+
"rewards/generated": -1.0779039859771729,
|
| 2109 |
+
"rewards/margins": 1.1664670705795288,
|
| 2110 |
+
"rewards/real": 0.08856303989887238,
|
| 2111 |
+
"step": 685
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.88,
|
| 2115 |
+
"grad_norm": 21.75,
|
| 2116 |
+
"learning_rate": 6.481481481481481e-08,
|
| 2117 |
+
"logits/generated": 0.20344486832618713,
|
| 2118 |
+
"logits/real": -0.2555938959121704,
|
| 2119 |
+
"logps/generated": -569.4146728515625,
|
| 2120 |
+
"logps/real": -463.9098205566406,
|
| 2121 |
+
"loss": 0.2581,
|
| 2122 |
+
"rewards/accuracies": 0.96875,
|
| 2123 |
+
"rewards/generated": -0.9887621998786926,
|
| 2124 |
+
"rewards/margins": 1.0963895320892334,
|
| 2125 |
+
"rewards/real": 0.10762734711170197,
|
| 2126 |
+
"step": 690
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 0.89,
|
| 2130 |
+
"grad_norm": 22.125,
|
| 2131 |
+
"learning_rate": 6.125356125356125e-08,
|
| 2132 |
+
"logits/generated": 0.23496928811073303,
|
| 2133 |
+
"logits/real": -0.2586161494255066,
|
| 2134 |
+
"logps/generated": -598.0418090820312,
|
| 2135 |
+
"logps/real": -467.5443420410156,
|
| 2136 |
+
"loss": 0.26,
|
| 2137 |
+
"rewards/accuracies": 0.949999988079071,
|
| 2138 |
+
"rewards/generated": -1.009071946144104,
|
| 2139 |
+
"rewards/margins": 1.0756361484527588,
|
| 2140 |
+
"rewards/real": 0.0665643960237503,
|
| 2141 |
+
"step": 695
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 0.9,
|
| 2145 |
+
"grad_norm": 22.5,
|
| 2146 |
+
"learning_rate": 5.7692307692307695e-08,
|
| 2147 |
+
"logits/generated": 0.25314944982528687,
|
| 2148 |
+
"logits/real": -0.2648155093193054,
|
| 2149 |
+
"logps/generated": -568.5360107421875,
|
| 2150 |
+
"logps/real": -475.3414001464844,
|
| 2151 |
+
"loss": 0.2678,
|
| 2152 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 2153 |
+
"rewards/generated": -1.0467432737350464,
|
| 2154 |
+
"rewards/margins": 1.0825011730194092,
|
| 2155 |
+
"rewards/real": 0.03575799986720085,
|
| 2156 |
+
"step": 700
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 0.9,
|
| 2160 |
+
"grad_norm": 22.25,
|
| 2161 |
+
"learning_rate": 5.413105413105413e-08,
|
| 2162 |
+
"logits/generated": 0.22809025645256042,
|
| 2163 |
+
"logits/real": -0.22145231068134308,
|
| 2164 |
+
"logps/generated": -562.1895141601562,
|
| 2165 |
+
"logps/real": -402.9983825683594,
|
| 2166 |
+
"loss": 0.2457,
|
| 2167 |
+
"rewards/accuracies": 0.949999988079071,
|
| 2168 |
+
"rewards/generated": -0.9787503480911255,
|
| 2169 |
+
"rewards/margins": 1.1118056774139404,
|
| 2170 |
+
"rewards/real": 0.13305525481700897,
|
| 2171 |
+
"step": 705
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 0.91,
|
| 2175 |
+
"grad_norm": 24.125,
|
| 2176 |
+
"learning_rate": 5.056980056980057e-08,
|
| 2177 |
+
"logits/generated": 0.2158997356891632,
|
| 2178 |
+
"logits/real": -0.33214133977890015,
|
| 2179 |
+
"logps/generated": -581.0538940429688,
|
| 2180 |
+
"logps/real": -487.0838928222656,
|
| 2181 |
+
"loss": 0.2588,
|
| 2182 |
+
"rewards/accuracies": 0.96875,
|
| 2183 |
+
"rewards/generated": -1.1220922470092773,
|
| 2184 |
+
"rewards/margins": 1.1539920568466187,
|
| 2185 |
+
"rewards/real": 0.03189963847398758,
|
| 2186 |
+
"step": 710
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.92,
|
| 2190 |
+
"grad_norm": 21.75,
|
| 2191 |
+
"learning_rate": 4.7008547008547005e-08,
|
| 2192 |
+
"logits/generated": 0.24275963008403778,
|
| 2193 |
+
"logits/real": -0.3125799894332886,
|
| 2194 |
+
"logps/generated": -556.0437622070312,
|
| 2195 |
+
"logps/real": -493.3427734375,
|
| 2196 |
+
"loss": 0.265,
|
| 2197 |
+
"rewards/accuracies": 0.918749988079071,
|
| 2198 |
+
"rewards/generated": -1.0702166557312012,
|
| 2199 |
+
"rewards/margins": 1.044668197631836,
|
| 2200 |
+
"rewards/real": -0.02554836869239807,
|
| 2201 |
+
"step": 715
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.92,
|
| 2205 |
+
"grad_norm": 21.625,
|
| 2206 |
+
"learning_rate": 4.3447293447293445e-08,
|
| 2207 |
+
"logits/generated": 0.21104946732521057,
|
| 2208 |
+
"logits/real": -0.21389278769493103,
|
| 2209 |
+
"logps/generated": -585.9505615234375,
|
| 2210 |
+
"logps/real": -407.6512756347656,
|
| 2211 |
+
"loss": 0.2583,
|
| 2212 |
+
"rewards/accuracies": 0.925000011920929,
|
| 2213 |
+
"rewards/generated": -1.045369029045105,
|
| 2214 |
+
"rewards/margins": 1.1055140495300293,
|
| 2215 |
+
"rewards/real": 0.06014498323202133,
|
| 2216 |
+
"step": 720
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 0.93,
|
| 2220 |
+
"grad_norm": 21.75,
|
| 2221 |
+
"learning_rate": 3.9886039886039886e-08,
|
| 2222 |
+
"logits/generated": 0.17947638034820557,
|
| 2223 |
+
"logits/real": -0.3292226195335388,
|
| 2224 |
+
"logps/generated": -536.8143310546875,
|
| 2225 |
+
"logps/real": -448.7561950683594,
|
| 2226 |
+
"loss": 0.2664,
|
| 2227 |
+
"rewards/accuracies": 0.925000011920929,
|
| 2228 |
+
"rewards/generated": -0.9618861079216003,
|
| 2229 |
+
"rewards/margins": 1.0169373750686646,
|
| 2230 |
+
"rewards/real": 0.05505116656422615,
|
| 2231 |
+
"step": 725
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"epoch": 0.93,
|
| 2235 |
+
"grad_norm": 20.875,
|
| 2236 |
+
"learning_rate": 3.632478632478633e-08,
|
| 2237 |
+
"logits/generated": 0.21649348735809326,
|
| 2238 |
+
"logits/real": -0.3055981993675232,
|
| 2239 |
+
"logps/generated": -581.56103515625,
|
| 2240 |
+
"logps/real": -459.6167907714844,
|
| 2241 |
+
"loss": 0.2434,
|
| 2242 |
+
"rewards/accuracies": 0.956250011920929,
|
| 2243 |
+
"rewards/generated": -1.0903055667877197,
|
| 2244 |
+
"rewards/margins": 1.1792160272598267,
|
| 2245 |
+
"rewards/real": 0.08891044557094574,
|
| 2246 |
+
"step": 730
|
| 2247 |
+
},
|
| 2248 |
+
{
|
| 2249 |
+
"epoch": 0.94,
|
| 2250 |
+
"grad_norm": 21.0,
|
| 2251 |
+
"learning_rate": 3.276353276353276e-08,
|
| 2252 |
+
"logits/generated": 0.218745619058609,
|
| 2253 |
+
"logits/real": -0.3472137749195099,
|
| 2254 |
+
"logps/generated": -565.79931640625,
|
| 2255 |
+
"logps/real": -489.8983459472656,
|
| 2256 |
+
"loss": 0.2563,
|
| 2257 |
+
"rewards/accuracies": 0.956250011920929,
|
| 2258 |
+
"rewards/generated": -1.093998908996582,
|
| 2259 |
+
"rewards/margins": 1.1649422645568848,
|
| 2260 |
+
"rewards/real": 0.0709434524178505,
|
| 2261 |
+
"step": 735
|
| 2262 |
+
},
|
| 2263 |
+
{
|
| 2264 |
+
"epoch": 0.95,
|
| 2265 |
+
"grad_norm": 22.375,
|
| 2266 |
+
"learning_rate": 2.92022792022792e-08,
|
| 2267 |
+
"logits/generated": 0.2501876950263977,
|
| 2268 |
+
"logits/real": -0.23009344935417175,
|
| 2269 |
+
"logps/generated": -569.3104248046875,
|
| 2270 |
+
"logps/real": -435.45599365234375,
|
| 2271 |
+
"loss": 0.242,
|
| 2272 |
+
"rewards/accuracies": 0.956250011920929,
|
| 2273 |
+
"rewards/generated": -1.0212743282318115,
|
| 2274 |
+
"rewards/margins": 1.118667721748352,
|
| 2275 |
+
"rewards/real": 0.09739341586828232,
|
| 2276 |
+
"step": 740
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 0.95,
|
| 2280 |
+
"grad_norm": 22.0,
|
| 2281 |
+
"learning_rate": 2.564102564102564e-08,
|
| 2282 |
+
"logits/generated": 0.2094719409942627,
|
| 2283 |
+
"logits/real": -0.2892032563686371,
|
| 2284 |
+
"logps/generated": -561.4392700195312,
|
| 2285 |
+
"logps/real": -422.3382263183594,
|
| 2286 |
+
"loss": 0.2481,
|
| 2287 |
+
"rewards/accuracies": 0.96875,
|
| 2288 |
+
"rewards/generated": -1.0572887659072876,
|
| 2289 |
+
"rewards/margins": 1.1870763301849365,
|
| 2290 |
+
"rewards/real": 0.1297876536846161,
|
| 2291 |
+
"step": 745
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.96,
|
| 2295 |
+
"grad_norm": 21.125,
|
| 2296 |
+
"learning_rate": 2.2079772079772077e-08,
|
| 2297 |
+
"logits/generated": 0.22928491234779358,
|
| 2298 |
+
"logits/real": -0.27726420760154724,
|
| 2299 |
+
"logps/generated": -603.4669799804688,
|
| 2300 |
+
"logps/real": -450.5899353027344,
|
| 2301 |
+
"loss": 0.2473,
|
| 2302 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 2303 |
+
"rewards/generated": -1.1500332355499268,
|
| 2304 |
+
"rewards/margins": 1.2057290077209473,
|
| 2305 |
+
"rewards/real": 0.055695801973342896,
|
| 2306 |
+
"step": 750
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.97,
|
| 2310 |
+
"grad_norm": 22.125,
|
| 2311 |
+
"learning_rate": 1.8518518518518518e-08,
|
| 2312 |
+
"logits/generated": 0.17330752313137054,
|
| 2313 |
+
"logits/real": -0.3336513638496399,
|
| 2314 |
+
"logps/generated": -548.7659912109375,
|
| 2315 |
+
"logps/real": -464.267578125,
|
| 2316 |
+
"loss": 0.2472,
|
| 2317 |
+
"rewards/accuracies": 0.9375,
|
| 2318 |
+
"rewards/generated": -1.0935777425765991,
|
| 2319 |
+
"rewards/margins": 1.205217957496643,
|
| 2320 |
+
"rewards/real": 0.11164028942584991,
|
| 2321 |
+
"step": 755
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 0.97,
|
| 2325 |
+
"grad_norm": 21.75,
|
| 2326 |
+
"learning_rate": 1.4957264957264956e-08,
|
| 2327 |
+
"logits/generated": 0.24052992463111877,
|
| 2328 |
+
"logits/real": -0.26754871010780334,
|
| 2329 |
+
"logps/generated": -572.6226806640625,
|
| 2330 |
+
"logps/real": -469.23388671875,
|
| 2331 |
+
"loss": 0.2562,
|
| 2332 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 2333 |
+
"rewards/generated": -0.9951506853103638,
|
| 2334 |
+
"rewards/margins": 1.117073655128479,
|
| 2335 |
+
"rewards/real": 0.12192295491695404,
|
| 2336 |
+
"step": 760
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 0.98,
|
| 2340 |
+
"grad_norm": 22.75,
|
| 2341 |
+
"learning_rate": 1.1396011396011397e-08,
|
| 2342 |
+
"logits/generated": 0.16937658190727234,
|
| 2343 |
+
"logits/real": -0.3556388020515442,
|
| 2344 |
+
"logps/generated": -551.8524169921875,
|
| 2345 |
+
"logps/real": -432.68316650390625,
|
| 2346 |
+
"loss": 0.2612,
|
| 2347 |
+
"rewards/accuracies": 0.9375,
|
| 2348 |
+
"rewards/generated": -1.0066879987716675,
|
| 2349 |
+
"rewards/margins": 1.067731261253357,
|
| 2350 |
+
"rewards/real": 0.06104331091046333,
|
| 2351 |
+
"step": 765
|
| 2352 |
+
},
|
| 2353 |
+
{
|
| 2354 |
+
"epoch": 0.99,
|
| 2355 |
+
"grad_norm": 22.625,
|
| 2356 |
+
"learning_rate": 7.834757834757834e-09,
|
| 2357 |
+
"logits/generated": 0.2987242341041565,
|
| 2358 |
+
"logits/real": -0.2289242297410965,
|
| 2359 |
+
"logps/generated": -591.0054931640625,
|
| 2360 |
+
"logps/real": -470.20758056640625,
|
| 2361 |
+
"loss": 0.2521,
|
| 2362 |
+
"rewards/accuracies": 0.949999988079071,
|
| 2363 |
+
"rewards/generated": -1.0826890468597412,
|
| 2364 |
+
"rewards/margins": 1.1511424779891968,
|
| 2365 |
+
"rewards/real": 0.06845370680093765,
|
| 2366 |
+
"step": 770
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 0.99,
|
| 2370 |
+
"grad_norm": 20.375,
|
| 2371 |
+
"learning_rate": 4.273504273504274e-09,
|
| 2372 |
+
"logits/generated": 0.2650791108608246,
|
| 2373 |
+
"logits/real": -0.2300834357738495,
|
| 2374 |
+
"logps/generated": -569.5623779296875,
|
| 2375 |
+
"logps/real": -417.6439514160156,
|
| 2376 |
+
"loss": 0.2514,
|
| 2377 |
+
"rewards/accuracies": 0.956250011920929,
|
| 2378 |
+
"rewards/generated": -1.0691241025924683,
|
| 2379 |
+
"rewards/margins": 1.168656826019287,
|
| 2380 |
+
"rewards/real": 0.09953261911869049,
|
| 2381 |
+
"step": 775
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 1.0,
|
| 2385 |
+
"grad_norm": 23.25,
|
| 2386 |
+
"learning_rate": 7.122507122507123e-10,
|
| 2387 |
+
"logits/generated": 0.22686879336833954,
|
| 2388 |
+
"logits/real": -0.32619014382362366,
|
| 2389 |
+
"logps/generated": -537.1260986328125,
|
| 2390 |
+
"logps/real": -499.892578125,
|
| 2391 |
+
"loss": 0.2663,
|
| 2392 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 2393 |
+
"rewards/generated": -0.9691260457038879,
|
| 2394 |
+
"rewards/margins": 1.0396109819412231,
|
| 2395 |
+
"rewards/real": 0.07048489153385162,
|
| 2396 |
+
"step": 780
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 1.0,
|
| 2400 |
+
"step": 781,
|
| 2401 |
+
"total_flos": 0.0,
|
| 2402 |
+
"train_loss": 0.31094988259020595,
|
| 2403 |
+
"train_runtime": 9866.2092,
|
| 2404 |
+
"train_samples_per_second": 5.068,
|
| 2405 |
+
"train_steps_per_second": 0.079
|
| 2406 |
+
}
|
| 2407 |
+
],
|
| 2408 |
+
"logging_steps": 5,
|
| 2409 |
+
"max_steps": 781,
|
| 2410 |
+
"num_input_tokens_seen": 0,
|
| 2411 |
+
"num_train_epochs": 1,
|
| 2412 |
+
"save_steps": 500,
|
| 2413 |
+
"total_flos": 0.0,
|
| 2414 |
+
"train_batch_size": 4,
|
| 2415 |
+
"trial_name": null,
|
| 2416 |
+
"trial_params": null
|
| 2417 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee29cb2e86a5c2ee5c31d447c0bca656972d759025a42e2ac2adbf89d9b75340
|
| 3 |
+
size 5048
|