Add checkpoint at step 7000
Browse files- checkpoints/checkpoint-step-7000/config.json +39 -0
- checkpoints/checkpoint-step-7000/generation_config.json +9 -0
- checkpoints/checkpoint-step-7000/merges.txt +0 -0
- checkpoints/checkpoint-step-7000/model.safetensors +3 -0
- checkpoints/checkpoint-step-7000/optimizer.pt +3 -0
- checkpoints/checkpoint-step-7000/rng_state.pth +3 -0
- checkpoints/checkpoint-step-7000/scaler.pt +3 -0
- checkpoints/checkpoint-step-7000/scheduler.pt +3 -0
- checkpoints/checkpoint-step-7000/special_tokens_map.json +12 -0
- checkpoints/checkpoint-step-7000/tokenizer.json +0 -0
- checkpoints/checkpoint-step-7000/tokenizer_config.json +21 -0
- checkpoints/checkpoint-step-7000/trainer_state.json +1294 -0
- checkpoints/checkpoint-step-7000/training_args.bin +3 -0
- checkpoints/checkpoint-step-7000/vocab.json +0 -0
checkpoints/checkpoint-step-7000/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"pad_token_id": 50256,
|
| 21 |
+
"reorder_and_upcast_attn": false,
|
| 22 |
+
"resid_pdrop": 0.1,
|
| 23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 24 |
+
"scale_attn_weights": true,
|
| 25 |
+
"summary_activation": null,
|
| 26 |
+
"summary_first_dropout": 0.1,
|
| 27 |
+
"summary_proj_to_labels": true,
|
| 28 |
+
"summary_type": "cls_index",
|
| 29 |
+
"summary_use_proj": true,
|
| 30 |
+
"task_specific_params": {
|
| 31 |
+
"text-generation": {
|
| 32 |
+
"do_sample": true,
|
| 33 |
+
"max_length": 50
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"transformers_version": "4.56.1",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
checkpoints/checkpoint-step-7000/generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
50256
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 50256,
|
| 8 |
+
"transformers_version": "4.56.1"
|
| 9 |
+
}
|
checkpoints/checkpoint-step-7000/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints/checkpoint-step-7000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:784db0bf710b7d2f5d21db4bd60e429a12dc79d0c719e202a19c550528b87388
|
| 3 |
+
size 497774208
|
checkpoints/checkpoint-step-7000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e014ded9d6c452586406e663e6e8d5de9894ed6a94b32a9d69f78f025778dc8
|
| 3 |
+
size 995644811
|
checkpoints/checkpoint-step-7000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16466d0333028c482007aea45193f9e1865b181a1c01becd71015d19745e9d94
|
| 3 |
+
size 14645
|
checkpoints/checkpoint-step-7000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a68ceae7165f4c88c8c4809f6ef958049408efc4229565cd13a1f7800728538
|
| 3 |
+
size 1383
|
checkpoints/checkpoint-step-7000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a19aa3c3e9956335a914b5072c3c48bed5d0544441ba0da1a5c689d79593efb
|
| 3 |
+
size 1465
|
checkpoints/checkpoint-step-7000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false
|
| 10 |
+
},
|
| 11 |
+
"unk_token": "<|endoftext|>"
|
| 12 |
+
}
|
checkpoints/checkpoint-step-7000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints/checkpoint-step-7000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
| 14 |
+
"clean_up_tokenization_spaces": false,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
| 16 |
+
"extra_special_tokens": {},
|
| 17 |
+
"model_max_length": 1024,
|
| 18 |
+
"pad_token": "<|endoftext|>",
|
| 19 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 20 |
+
"unk_token": "<|endoftext|>"
|
| 21 |
+
}
|
checkpoints/checkpoint-step-7000/trainer_state.json
ADDED
|
@@ -0,0 +1,1294 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6800,
|
| 3 |
+
"best_metric": 0.029951954260468483,
|
| 4 |
+
"best_model_checkpoint": "checkpoints/checkpoint-6800",
|
| 5 |
+
"epoch": 1.9047619047619047,
|
| 6 |
+
"eval_steps": 200,
|
| 7 |
+
"global_step": 7000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.013605442176870748,
|
| 14 |
+
"grad_norm": 4.6472649574279785,
|
| 15 |
+
"learning_rate": 4.9000000000000005e-06,
|
| 16 |
+
"loss": 2.8759,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.027210884353741496,
|
| 21 |
+
"grad_norm": 3.089015245437622,
|
| 22 |
+
"learning_rate": 9.900000000000002e-06,
|
| 23 |
+
"loss": 1.7949,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.04081632653061224,
|
| 28 |
+
"grad_norm": 2.3650102615356445,
|
| 29 |
+
"learning_rate": 1.49e-05,
|
| 30 |
+
"loss": 0.6938,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.05442176870748299,
|
| 35 |
+
"grad_norm": 1.4423813819885254,
|
| 36 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 37 |
+
"loss": 0.2987,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05442176870748299,
|
| 42 |
+
"eval_loss": 0.18158380687236786,
|
| 43 |
+
"eval_runtime": 10.2028,
|
| 44 |
+
"eval_samples_per_second": 58.808,
|
| 45 |
+
"eval_steps_per_second": 7.351,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.06802721088435375,
|
| 50 |
+
"grad_norm": 0.9667473435401917,
|
| 51 |
+
"learning_rate": 2.4900000000000002e-05,
|
| 52 |
+
"loss": 0.2009,
|
| 53 |
+
"step": 250
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.08163265306122448,
|
| 57 |
+
"grad_norm": 1.2099648714065552,
|
| 58 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 59 |
+
"loss": 0.1613,
|
| 60 |
+
"step": 300
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.09523809523809523,
|
| 64 |
+
"grad_norm": 1.036044955253601,
|
| 65 |
+
"learning_rate": 3.49e-05,
|
| 66 |
+
"loss": 0.1397,
|
| 67 |
+
"step": 350
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.10884353741496598,
|
| 71 |
+
"grad_norm": 0.7627539038658142,
|
| 72 |
+
"learning_rate": 3.99e-05,
|
| 73 |
+
"loss": 0.1236,
|
| 74 |
+
"step": 400
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.10884353741496598,
|
| 78 |
+
"eval_loss": 0.09973898530006409,
|
| 79 |
+
"eval_runtime": 10.2069,
|
| 80 |
+
"eval_samples_per_second": 58.784,
|
| 81 |
+
"eval_steps_per_second": 7.348,
|
| 82 |
+
"step": 400
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.12244897959183673,
|
| 86 |
+
"grad_norm": 1.1398504972457886,
|
| 87 |
+
"learning_rate": 4.49e-05,
|
| 88 |
+
"loss": 0.1149,
|
| 89 |
+
"step": 450
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.1360544217687075,
|
| 93 |
+
"grad_norm": 0.7201138734817505,
|
| 94 |
+
"learning_rate": 4.99e-05,
|
| 95 |
+
"loss": 0.1004,
|
| 96 |
+
"step": 500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.14965986394557823,
|
| 100 |
+
"grad_norm": 0.6430326700210571,
|
| 101 |
+
"learning_rate": 4.9642335766423356e-05,
|
| 102 |
+
"loss": 0.0926,
|
| 103 |
+
"step": 550
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.16326530612244897,
|
| 107 |
+
"grad_norm": 0.8617527484893799,
|
| 108 |
+
"learning_rate": 4.9277372262773724e-05,
|
| 109 |
+
"loss": 0.0876,
|
| 110 |
+
"step": 600
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.16326530612244897,
|
| 114 |
+
"eval_loss": 0.06922342628240585,
|
| 115 |
+
"eval_runtime": 10.155,
|
| 116 |
+
"eval_samples_per_second": 59.084,
|
| 117 |
+
"eval_steps_per_second": 7.386,
|
| 118 |
+
"step": 600
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.17687074829931973,
|
| 122 |
+
"grad_norm": 0.7790252566337585,
|
| 123 |
+
"learning_rate": 4.891240875912409e-05,
|
| 124 |
+
"loss": 0.0813,
|
| 125 |
+
"step": 650
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.19047619047619047,
|
| 129 |
+
"grad_norm": 0.6242516040802002,
|
| 130 |
+
"learning_rate": 4.854744525547445e-05,
|
| 131 |
+
"loss": 0.0778,
|
| 132 |
+
"step": 700
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.20408163265306123,
|
| 136 |
+
"grad_norm": 0.5622245073318481,
|
| 137 |
+
"learning_rate": 4.818248175182482e-05,
|
| 138 |
+
"loss": 0.0733,
|
| 139 |
+
"step": 750
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.21768707482993196,
|
| 143 |
+
"grad_norm": 0.6843573451042175,
|
| 144 |
+
"learning_rate": 4.781751824817519e-05,
|
| 145 |
+
"loss": 0.0703,
|
| 146 |
+
"step": 800
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.21768707482993196,
|
| 150 |
+
"eval_loss": 0.055413372814655304,
|
| 151 |
+
"eval_runtime": 10.1991,
|
| 152 |
+
"eval_samples_per_second": 58.829,
|
| 153 |
+
"eval_steps_per_second": 7.354,
|
| 154 |
+
"step": 800
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.23129251700680273,
|
| 158 |
+
"grad_norm": 0.611773669719696,
|
| 159 |
+
"learning_rate": 4.745255474452555e-05,
|
| 160 |
+
"loss": 0.0668,
|
| 161 |
+
"step": 850
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.24489795918367346,
|
| 165 |
+
"grad_norm": 0.622644305229187,
|
| 166 |
+
"learning_rate": 4.708759124087592e-05,
|
| 167 |
+
"loss": 0.0649,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.2585034013605442,
|
| 172 |
+
"grad_norm": 0.4426730275154114,
|
| 173 |
+
"learning_rate": 4.6722627737226286e-05,
|
| 174 |
+
"loss": 0.0637,
|
| 175 |
+
"step": 950
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.272108843537415,
|
| 179 |
+
"grad_norm": 0.8717305660247803,
|
| 180 |
+
"learning_rate": 4.635766423357664e-05,
|
| 181 |
+
"loss": 0.0601,
|
| 182 |
+
"step": 1000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.272108843537415,
|
| 186 |
+
"eval_loss": 0.05313626304268837,
|
| 187 |
+
"eval_runtime": 10.1979,
|
| 188 |
+
"eval_samples_per_second": 58.836,
|
| 189 |
+
"eval_steps_per_second": 7.354,
|
| 190 |
+
"step": 1000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.2857142857142857,
|
| 194 |
+
"grad_norm": 0.5126848220825195,
|
| 195 |
+
"learning_rate": 4.599270072992701e-05,
|
| 196 |
+
"loss": 0.0578,
|
| 197 |
+
"step": 1050
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.29931972789115646,
|
| 201 |
+
"grad_norm": 0.5319710373878479,
|
| 202 |
+
"learning_rate": 4.5627737226277376e-05,
|
| 203 |
+
"loss": 0.0574,
|
| 204 |
+
"step": 1100
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.3129251700680272,
|
| 208 |
+
"grad_norm": 0.7722771167755127,
|
| 209 |
+
"learning_rate": 4.526277372262774e-05,
|
| 210 |
+
"loss": 0.0564,
|
| 211 |
+
"step": 1150
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.32653061224489793,
|
| 215 |
+
"grad_norm": 0.4977850317955017,
|
| 216 |
+
"learning_rate": 4.4897810218978105e-05,
|
| 217 |
+
"loss": 0.0553,
|
| 218 |
+
"step": 1200
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.32653061224489793,
|
| 222 |
+
"eval_loss": 0.0452888123691082,
|
| 223 |
+
"eval_runtime": 10.1684,
|
| 224 |
+
"eval_samples_per_second": 59.006,
|
| 225 |
+
"eval_steps_per_second": 7.376,
|
| 226 |
+
"step": 1200
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3401360544217687,
|
| 230 |
+
"grad_norm": 0.5616284608840942,
|
| 231 |
+
"learning_rate": 4.4532846715328466e-05,
|
| 232 |
+
"loss": 0.0529,
|
| 233 |
+
"step": 1250
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.35374149659863946,
|
| 237 |
+
"grad_norm": 0.41897886991500854,
|
| 238 |
+
"learning_rate": 4.4167883211678834e-05,
|
| 239 |
+
"loss": 0.0527,
|
| 240 |
+
"step": 1300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3673469387755102,
|
| 244 |
+
"grad_norm": 0.526966392993927,
|
| 245 |
+
"learning_rate": 4.38029197080292e-05,
|
| 246 |
+
"loss": 0.0518,
|
| 247 |
+
"step": 1350
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.38095238095238093,
|
| 251 |
+
"grad_norm": 0.7886810898780823,
|
| 252 |
+
"learning_rate": 4.343795620437956e-05,
|
| 253 |
+
"loss": 0.0514,
|
| 254 |
+
"step": 1400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.38095238095238093,
|
| 258 |
+
"eval_loss": 0.042705778032541275,
|
| 259 |
+
"eval_runtime": 10.1958,
|
| 260 |
+
"eval_samples_per_second": 58.848,
|
| 261 |
+
"eval_steps_per_second": 7.356,
|
| 262 |
+
"step": 1400
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3945578231292517,
|
| 266 |
+
"grad_norm": 0.38092342019081116,
|
| 267 |
+
"learning_rate": 4.307299270072993e-05,
|
| 268 |
+
"loss": 0.0507,
|
| 269 |
+
"step": 1450
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.40816326530612246,
|
| 273 |
+
"grad_norm": 0.47930657863616943,
|
| 274 |
+
"learning_rate": 4.27080291970803e-05,
|
| 275 |
+
"loss": 0.0517,
|
| 276 |
+
"step": 1500
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4217687074829932,
|
| 280 |
+
"grad_norm": 0.529920220375061,
|
| 281 |
+
"learning_rate": 4.234306569343066e-05,
|
| 282 |
+
"loss": 0.0494,
|
| 283 |
+
"step": 1550
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.43537414965986393,
|
| 287 |
+
"grad_norm": 0.5892526507377625,
|
| 288 |
+
"learning_rate": 4.197810218978102e-05,
|
| 289 |
+
"loss": 0.0473,
|
| 290 |
+
"step": 1600
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.43537414965986393,
|
| 294 |
+
"eval_loss": 0.04030081257224083,
|
| 295 |
+
"eval_runtime": 10.1867,
|
| 296 |
+
"eval_samples_per_second": 58.9,
|
| 297 |
+
"eval_steps_per_second": 7.363,
|
| 298 |
+
"step": 1600
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.4489795918367347,
|
| 302 |
+
"grad_norm": 0.8671649098396301,
|
| 303 |
+
"learning_rate": 4.161313868613139e-05,
|
| 304 |
+
"loss": 0.047,
|
| 305 |
+
"step": 1650
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.46258503401360546,
|
| 309 |
+
"grad_norm": 0.6588522791862488,
|
| 310 |
+
"learning_rate": 4.124817518248175e-05,
|
| 311 |
+
"loss": 0.0481,
|
| 312 |
+
"step": 1700
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.47619047619047616,
|
| 316 |
+
"grad_norm": 0.502729594707489,
|
| 317 |
+
"learning_rate": 4.088321167883212e-05,
|
| 318 |
+
"loss": 0.0456,
|
| 319 |
+
"step": 1750
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.4897959183673469,
|
| 323 |
+
"grad_norm": 0.5965167284011841,
|
| 324 |
+
"learning_rate": 4.0518248175182486e-05,
|
| 325 |
+
"loss": 0.0463,
|
| 326 |
+
"step": 1800
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.4897959183673469,
|
| 330 |
+
"eval_loss": 0.040413301438093185,
|
| 331 |
+
"eval_runtime": 10.2208,
|
| 332 |
+
"eval_samples_per_second": 58.704,
|
| 333 |
+
"eval_steps_per_second": 7.338,
|
| 334 |
+
"step": 1800
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.5034013605442177,
|
| 338 |
+
"grad_norm": 0.38793548941612244,
|
| 339 |
+
"learning_rate": 4.015328467153285e-05,
|
| 340 |
+
"loss": 0.0451,
|
| 341 |
+
"step": 1850
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 0.5170068027210885,
|
| 345 |
+
"grad_norm": 0.3949367105960846,
|
| 346 |
+
"learning_rate": 3.9788321167883215e-05,
|
| 347 |
+
"loss": 0.0468,
|
| 348 |
+
"step": 1900
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.5306122448979592,
|
| 352 |
+
"grad_norm": 0.5778154134750366,
|
| 353 |
+
"learning_rate": 3.9423357664233576e-05,
|
| 354 |
+
"loss": 0.0454,
|
| 355 |
+
"step": 1950
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.54421768707483,
|
| 359 |
+
"grad_norm": 0.3722288906574249,
|
| 360 |
+
"learning_rate": 3.9058394160583944e-05,
|
| 361 |
+
"loss": 0.0461,
|
| 362 |
+
"step": 2000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.54421768707483,
|
| 366 |
+
"eval_loss": 0.03860794007778168,
|
| 367 |
+
"eval_runtime": 10.1795,
|
| 368 |
+
"eval_samples_per_second": 58.942,
|
| 369 |
+
"eval_steps_per_second": 7.368,
|
| 370 |
+
"step": 2000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.5578231292517006,
|
| 374 |
+
"grad_norm": 0.5253990292549133,
|
| 375 |
+
"learning_rate": 3.869343065693431e-05,
|
| 376 |
+
"loss": 0.0445,
|
| 377 |
+
"step": 2050
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.5714285714285714,
|
| 381 |
+
"grad_norm": 0.41715025901794434,
|
| 382 |
+
"learning_rate": 3.832846715328467e-05,
|
| 383 |
+
"loss": 0.042,
|
| 384 |
+
"step": 2100
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.5850340136054422,
|
| 388 |
+
"grad_norm": 0.567323625087738,
|
| 389 |
+
"learning_rate": 3.796350364963504e-05,
|
| 390 |
+
"loss": 0.0423,
|
| 391 |
+
"step": 2150
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.5986394557823129,
|
| 395 |
+
"grad_norm": 0.6290739178657532,
|
| 396 |
+
"learning_rate": 3.759854014598541e-05,
|
| 397 |
+
"loss": 0.0428,
|
| 398 |
+
"step": 2200
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.5986394557823129,
|
| 402 |
+
"eval_loss": 0.037571050226688385,
|
| 403 |
+
"eval_runtime": 10.1846,
|
| 404 |
+
"eval_samples_per_second": 58.913,
|
| 405 |
+
"eval_steps_per_second": 7.364,
|
| 406 |
+
"step": 2200
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.6122448979591837,
|
| 410 |
+
"grad_norm": 0.40591228008270264,
|
| 411 |
+
"learning_rate": 3.723357664233576e-05,
|
| 412 |
+
"loss": 0.043,
|
| 413 |
+
"step": 2250
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.6258503401360545,
|
| 417 |
+
"grad_norm": 0.45446255803108215,
|
| 418 |
+
"learning_rate": 3.686861313868613e-05,
|
| 419 |
+
"loss": 0.0423,
|
| 420 |
+
"step": 2300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.6394557823129252,
|
| 424 |
+
"grad_norm": 0.41105887293815613,
|
| 425 |
+
"learning_rate": 3.65036496350365e-05,
|
| 426 |
+
"loss": 0.0413,
|
| 427 |
+
"step": 2350
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.6530612244897959,
|
| 431 |
+
"grad_norm": 0.4538460671901703,
|
| 432 |
+
"learning_rate": 3.613868613138686e-05,
|
| 433 |
+
"loss": 0.0428,
|
| 434 |
+
"step": 2400
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.6530612244897959,
|
| 438 |
+
"eval_loss": 0.036736100912094116,
|
| 439 |
+
"eval_runtime": 10.1735,
|
| 440 |
+
"eval_samples_per_second": 58.977,
|
| 441 |
+
"eval_steps_per_second": 7.372,
|
| 442 |
+
"step": 2400
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.6666666666666666,
|
| 446 |
+
"grad_norm": 0.42337286472320557,
|
| 447 |
+
"learning_rate": 3.577372262773723e-05,
|
| 448 |
+
"loss": 0.0406,
|
| 449 |
+
"step": 2450
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.6802721088435374,
|
| 453 |
+
"grad_norm": 0.3924243450164795,
|
| 454 |
+
"learning_rate": 3.5408759124087596e-05,
|
| 455 |
+
"loss": 0.0412,
|
| 456 |
+
"step": 2500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.6938775510204082,
|
| 460 |
+
"grad_norm": 0.3243819773197174,
|
| 461 |
+
"learning_rate": 3.504379562043796e-05,
|
| 462 |
+
"loss": 0.0416,
|
| 463 |
+
"step": 2550
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.7074829931972789,
|
| 467 |
+
"grad_norm": 0.3431473970413208,
|
| 468 |
+
"learning_rate": 3.4678832116788325e-05,
|
| 469 |
+
"loss": 0.0395,
|
| 470 |
+
"step": 2600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.7074829931972789,
|
| 474 |
+
"eval_loss": 0.035297442227602005,
|
| 475 |
+
"eval_runtime": 10.181,
|
| 476 |
+
"eval_samples_per_second": 58.933,
|
| 477 |
+
"eval_steps_per_second": 7.367,
|
| 478 |
+
"step": 2600
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.7210884353741497,
|
| 482 |
+
"grad_norm": 0.4607592821121216,
|
| 483 |
+
"learning_rate": 3.4313868613138686e-05,
|
| 484 |
+
"loss": 0.0392,
|
| 485 |
+
"step": 2650
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7346938775510204,
|
| 489 |
+
"grad_norm": 0.5089300870895386,
|
| 490 |
+
"learning_rate": 3.3948905109489054e-05,
|
| 491 |
+
"loss": 0.0409,
|
| 492 |
+
"step": 2700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7482993197278912,
|
| 496 |
+
"grad_norm": 0.42247381806373596,
|
| 497 |
+
"learning_rate": 3.358394160583942e-05,
|
| 498 |
+
"loss": 0.041,
|
| 499 |
+
"step": 2750
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7619047619047619,
|
| 503 |
+
"grad_norm": 0.5103944540023804,
|
| 504 |
+
"learning_rate": 3.321897810218978e-05,
|
| 505 |
+
"loss": 0.0396,
|
| 506 |
+
"step": 2800
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7619047619047619,
|
| 510 |
+
"eval_loss": 0.03508320823311806,
|
| 511 |
+
"eval_runtime": 10.1991,
|
| 512 |
+
"eval_samples_per_second": 58.829,
|
| 513 |
+
"eval_steps_per_second": 7.354,
|
| 514 |
+
"step": 2800
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7755102040816326,
|
| 518 |
+
"grad_norm": 0.4671725928783417,
|
| 519 |
+
"learning_rate": 3.2854014598540144e-05,
|
| 520 |
+
"loss": 0.0401,
|
| 521 |
+
"step": 2850
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7891156462585034,
|
| 525 |
+
"grad_norm": 0.39482733607292175,
|
| 526 |
+
"learning_rate": 3.248905109489051e-05,
|
| 527 |
+
"loss": 0.0413,
|
| 528 |
+
"step": 2900
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.8027210884353742,
|
| 532 |
+
"grad_norm": 0.6017800569534302,
|
| 533 |
+
"learning_rate": 3.212408759124087e-05,
|
| 534 |
+
"loss": 0.039,
|
| 535 |
+
"step": 2950
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8163265306122449,
|
| 539 |
+
"grad_norm": 0.3402301073074341,
|
| 540 |
+
"learning_rate": 3.175912408759124e-05,
|
| 541 |
+
"loss": 0.0388,
|
| 542 |
+
"step": 3000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8163265306122449,
|
| 546 |
+
"eval_loss": 0.03466026484966278,
|
| 547 |
+
"eval_runtime": 10.1801,
|
| 548 |
+
"eval_samples_per_second": 58.938,
|
| 549 |
+
"eval_steps_per_second": 7.367,
|
| 550 |
+
"step": 3000
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.8299319727891157,
|
| 554 |
+
"grad_norm": 0.31006762385368347,
|
| 555 |
+
"learning_rate": 3.139416058394161e-05,
|
| 556 |
+
"loss": 0.0372,
|
| 557 |
+
"step": 3050
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.8435374149659864,
|
| 561 |
+
"grad_norm": 0.6701622605323792,
|
| 562 |
+
"learning_rate": 3.102919708029197e-05,
|
| 563 |
+
"loss": 0.0393,
|
| 564 |
+
"step": 3100
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.8571428571428571,
|
| 568 |
+
"grad_norm": 0.398631751537323,
|
| 569 |
+
"learning_rate": 3.066423357664234e-05,
|
| 570 |
+
"loss": 0.0396,
|
| 571 |
+
"step": 3150
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.8707482993197279,
|
| 575 |
+
"grad_norm": 0.3951578140258789,
|
| 576 |
+
"learning_rate": 3.0299270072992703e-05,
|
| 577 |
+
"loss": 0.0383,
|
| 578 |
+
"step": 3200
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.8707482993197279,
|
| 582 |
+
"eval_loss": 0.03498370572924614,
|
| 583 |
+
"eval_runtime": 10.1869,
|
| 584 |
+
"eval_samples_per_second": 58.899,
|
| 585 |
+
"eval_steps_per_second": 7.362,
|
| 586 |
+
"step": 3200
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.8843537414965986,
|
| 590 |
+
"grad_norm": 0.45307376980781555,
|
| 591 |
+
"learning_rate": 2.9934306569343067e-05,
|
| 592 |
+
"loss": 0.039,
|
| 593 |
+
"step": 3250
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.8979591836734694,
|
| 597 |
+
"grad_norm": 0.46504005789756775,
|
| 598 |
+
"learning_rate": 2.9569343065693432e-05,
|
| 599 |
+
"loss": 0.0372,
|
| 600 |
+
"step": 3300
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.9115646258503401,
|
| 604 |
+
"grad_norm": 0.3962818682193756,
|
| 605 |
+
"learning_rate": 2.92043795620438e-05,
|
| 606 |
+
"loss": 0.0388,
|
| 607 |
+
"step": 3350
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.9251700680272109,
|
| 611 |
+
"grad_norm": 0.5040358901023865,
|
| 612 |
+
"learning_rate": 2.8839416058394164e-05,
|
| 613 |
+
"loss": 0.038,
|
| 614 |
+
"step": 3400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.9251700680272109,
|
| 618 |
+
"eval_loss": 0.03590795397758484,
|
| 619 |
+
"eval_runtime": 10.1999,
|
| 620 |
+
"eval_samples_per_second": 58.824,
|
| 621 |
+
"eval_steps_per_second": 7.353,
|
| 622 |
+
"step": 3400
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.9387755102040817,
|
| 626 |
+
"grad_norm": 0.8045864701271057,
|
| 627 |
+
"learning_rate": 2.847445255474453e-05,
|
| 628 |
+
"loss": 0.0364,
|
| 629 |
+
"step": 3450
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.9523809523809523,
|
| 633 |
+
"grad_norm": 0.9472477436065674,
|
| 634 |
+
"learning_rate": 2.810948905109489e-05,
|
| 635 |
+
"loss": 0.0365,
|
| 636 |
+
"step": 3500
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.9659863945578231,
|
| 640 |
+
"grad_norm": 0.5935471057891846,
|
| 641 |
+
"learning_rate": 2.7744525547445254e-05,
|
| 642 |
+
"loss": 0.0358,
|
| 643 |
+
"step": 3550
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 0.9795918367346939,
|
| 647 |
+
"grad_norm": 0.42011234164237976,
|
| 648 |
+
"learning_rate": 2.737956204379562e-05,
|
| 649 |
+
"loss": 0.0371,
|
| 650 |
+
"step": 3600
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.9795918367346939,
|
| 654 |
+
"eval_loss": 0.034322191029787064,
|
| 655 |
+
"eval_runtime": 10.2002,
|
| 656 |
+
"eval_samples_per_second": 58.823,
|
| 657 |
+
"eval_steps_per_second": 7.353,
|
| 658 |
+
"step": 3600
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.9931972789115646,
|
| 662 |
+
"grad_norm": 0.757950484752655,
|
| 663 |
+
"learning_rate": 2.7014598540145987e-05,
|
| 664 |
+
"loss": 0.0367,
|
| 665 |
+
"step": 3650
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 1.0068027210884354,
|
| 669 |
+
"grad_norm": 0.37888622283935547,
|
| 670 |
+
"learning_rate": 2.664963503649635e-05,
|
| 671 |
+
"loss": 0.0376,
|
| 672 |
+
"step": 3700
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.0204081632653061,
|
| 676 |
+
"grad_norm": 0.4423586428165436,
|
| 677 |
+
"learning_rate": 2.6284671532846716e-05,
|
| 678 |
+
"loss": 0.0363,
|
| 679 |
+
"step": 3750
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 1.034013605442177,
|
| 683 |
+
"grad_norm": 0.526573896408081,
|
| 684 |
+
"learning_rate": 2.591970802919708e-05,
|
| 685 |
+
"loss": 0.0364,
|
| 686 |
+
"step": 3800
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.034013605442177,
|
| 690 |
+
"eval_loss": 0.03279593959450722,
|
| 691 |
+
"eval_runtime": 10.1819,
|
| 692 |
+
"eval_samples_per_second": 58.928,
|
| 693 |
+
"eval_steps_per_second": 7.366,
|
| 694 |
+
"step": 3800
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 1.0476190476190477,
|
| 698 |
+
"grad_norm": 0.32934117317199707,
|
| 699 |
+
"learning_rate": 2.555474452554745e-05,
|
| 700 |
+
"loss": 0.0359,
|
| 701 |
+
"step": 3850
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 1.0612244897959184,
|
| 705 |
+
"grad_norm": 0.33360642194747925,
|
| 706 |
+
"learning_rate": 2.5189781021897813e-05,
|
| 707 |
+
"loss": 0.0376,
|
| 708 |
+
"step": 3900
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.0748299319727892,
|
| 712 |
+
"grad_norm": 0.6703974604606628,
|
| 713 |
+
"learning_rate": 2.4824817518248174e-05,
|
| 714 |
+
"loss": 0.0369,
|
| 715 |
+
"step": 3950
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 1.08843537414966,
|
| 719 |
+
"grad_norm": 0.3145996034145355,
|
| 720 |
+
"learning_rate": 2.4459854014598542e-05,
|
| 721 |
+
"loss": 0.0372,
|
| 722 |
+
"step": 4000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.08843537414966,
|
| 726 |
+
"eval_loss": 0.0330994687974453,
|
| 727 |
+
"eval_runtime": 10.1837,
|
| 728 |
+
"eval_samples_per_second": 58.918,
|
| 729 |
+
"eval_steps_per_second": 7.365,
|
| 730 |
+
"step": 4000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.1020408163265305,
|
| 734 |
+
"grad_norm": 0.5301318168640137,
|
| 735 |
+
"learning_rate": 2.4094890510948906e-05,
|
| 736 |
+
"loss": 0.0365,
|
| 737 |
+
"step": 4050
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.1156462585034013,
|
| 741 |
+
"grad_norm": 0.7140094637870789,
|
| 742 |
+
"learning_rate": 2.372992700729927e-05,
|
| 743 |
+
"loss": 0.0358,
|
| 744 |
+
"step": 4100
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.129251700680272,
|
| 748 |
+
"grad_norm": 0.3584352433681488,
|
| 749 |
+
"learning_rate": 2.3364963503649635e-05,
|
| 750 |
+
"loss": 0.0356,
|
| 751 |
+
"step": 4150
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.1428571428571428,
|
| 755 |
+
"grad_norm": 0.32996875047683716,
|
| 756 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 757 |
+
"loss": 0.0363,
|
| 758 |
+
"step": 4200
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.1428571428571428,
|
| 762 |
+
"eval_loss": 0.03242386505007744,
|
| 763 |
+
"eval_runtime": 10.2256,
|
| 764 |
+
"eval_samples_per_second": 58.676,
|
| 765 |
+
"eval_steps_per_second": 7.335,
|
| 766 |
+
"step": 4200
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.1564625850340136,
|
| 770 |
+
"grad_norm": 0.3712214231491089,
|
| 771 |
+
"learning_rate": 2.2635036496350365e-05,
|
| 772 |
+
"loss": 0.0355,
|
| 773 |
+
"step": 4250
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.1700680272108843,
|
| 777 |
+
"grad_norm": 0.4751546084880829,
|
| 778 |
+
"learning_rate": 2.227007299270073e-05,
|
| 779 |
+
"loss": 0.0347,
|
| 780 |
+
"step": 4300
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.183673469387755,
|
| 784 |
+
"grad_norm": 0.49500930309295654,
|
| 785 |
+
"learning_rate": 2.1905109489051097e-05,
|
| 786 |
+
"loss": 0.0362,
|
| 787 |
+
"step": 4350
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.1972789115646258,
|
| 791 |
+
"grad_norm": 0.4981195032596588,
|
| 792 |
+
"learning_rate": 2.154014598540146e-05,
|
| 793 |
+
"loss": 0.0351,
|
| 794 |
+
"step": 4400
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.1972789115646258,
|
| 798 |
+
"eval_loss": 0.033444974571466446,
|
| 799 |
+
"eval_runtime": 10.2088,
|
| 800 |
+
"eval_samples_per_second": 58.773,
|
| 801 |
+
"eval_steps_per_second": 7.347,
|
| 802 |
+
"step": 4400
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 1.2108843537414966,
|
| 806 |
+
"grad_norm": 0.30703791975975037,
|
| 807 |
+
"learning_rate": 2.1175182481751826e-05,
|
| 808 |
+
"loss": 0.0348,
|
| 809 |
+
"step": 4450
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.2244897959183674,
|
| 813 |
+
"grad_norm": 0.40786242485046387,
|
| 814 |
+
"learning_rate": 2.081021897810219e-05,
|
| 815 |
+
"loss": 0.0348,
|
| 816 |
+
"step": 4500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.2380952380952381,
|
| 820 |
+
"grad_norm": 0.33420053124427795,
|
| 821 |
+
"learning_rate": 2.044525547445256e-05,
|
| 822 |
+
"loss": 0.0349,
|
| 823 |
+
"step": 4550
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.251700680272109,
|
| 827 |
+
"grad_norm": 0.39799734950065613,
|
| 828 |
+
"learning_rate": 2.008029197080292e-05,
|
| 829 |
+
"loss": 0.0347,
|
| 830 |
+
"step": 4600
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.251700680272109,
|
| 834 |
+
"eval_loss": 0.03165949881076813,
|
| 835 |
+
"eval_runtime": 10.4306,
|
| 836 |
+
"eval_samples_per_second": 57.523,
|
| 837 |
+
"eval_steps_per_second": 7.19,
|
| 838 |
+
"step": 4600
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 1.2653061224489797,
|
| 842 |
+
"grad_norm": 0.5049504637718201,
|
| 843 |
+
"learning_rate": 1.9715328467153284e-05,
|
| 844 |
+
"loss": 0.0358,
|
| 845 |
+
"step": 4650
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.2789115646258504,
|
| 849 |
+
"grad_norm": 0.41933709383010864,
|
| 850 |
+
"learning_rate": 1.9350364963503652e-05,
|
| 851 |
+
"loss": 0.0338,
|
| 852 |
+
"step": 4700
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.2925170068027212,
|
| 856 |
+
"grad_norm": 0.6357618570327759,
|
| 857 |
+
"learning_rate": 1.8985401459854017e-05,
|
| 858 |
+
"loss": 0.0351,
|
| 859 |
+
"step": 4750
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.306122448979592,
|
| 863 |
+
"grad_norm": 0.34428149461746216,
|
| 864 |
+
"learning_rate": 1.862043795620438e-05,
|
| 865 |
+
"loss": 0.0342,
|
| 866 |
+
"step": 4800
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.306122448979592,
|
| 870 |
+
"eval_loss": 0.03147235885262489,
|
| 871 |
+
"eval_runtime": 10.2354,
|
| 872 |
+
"eval_samples_per_second": 58.62,
|
| 873 |
+
"eval_steps_per_second": 7.327,
|
| 874 |
+
"step": 4800
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 1.3197278911564627,
|
| 878 |
+
"grad_norm": 0.3444003760814667,
|
| 879 |
+
"learning_rate": 1.8255474452554746e-05,
|
| 880 |
+
"loss": 0.035,
|
| 881 |
+
"step": 4850
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 1.3333333333333333,
|
| 885 |
+
"grad_norm": 0.3341350853443146,
|
| 886 |
+
"learning_rate": 1.789051094890511e-05,
|
| 887 |
+
"loss": 0.0352,
|
| 888 |
+
"step": 4900
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 1.346938775510204,
|
| 892 |
+
"grad_norm": 0.3407929241657257,
|
| 893 |
+
"learning_rate": 1.7525547445255475e-05,
|
| 894 |
+
"loss": 0.0343,
|
| 895 |
+
"step": 4950
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 1.3605442176870748,
|
| 899 |
+
"grad_norm": 0.5530718564987183,
|
| 900 |
+
"learning_rate": 1.716058394160584e-05,
|
| 901 |
+
"loss": 0.0344,
|
| 902 |
+
"step": 5000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.3605442176870748,
|
| 906 |
+
"eval_loss": 0.031369421631097794,
|
| 907 |
+
"eval_runtime": 10.1936,
|
| 908 |
+
"eval_samples_per_second": 58.86,
|
| 909 |
+
"eval_steps_per_second": 7.358,
|
| 910 |
+
"step": 5000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 1.3741496598639455,
|
| 914 |
+
"grad_norm": 0.43771809339523315,
|
| 915 |
+
"learning_rate": 1.6795620437956207e-05,
|
| 916 |
+
"loss": 0.0359,
|
| 917 |
+
"step": 5050
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.3877551020408163,
|
| 921 |
+
"grad_norm": 0.3095148503780365,
|
| 922 |
+
"learning_rate": 1.643065693430657e-05,
|
| 923 |
+
"loss": 0.0339,
|
| 924 |
+
"step": 5100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.401360544217687,
|
| 928 |
+
"grad_norm": 0.35238730907440186,
|
| 929 |
+
"learning_rate": 1.6065693430656936e-05,
|
| 930 |
+
"loss": 0.0344,
|
| 931 |
+
"step": 5150
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 1.4149659863945578,
|
| 935 |
+
"grad_norm": 0.5534179210662842,
|
| 936 |
+
"learning_rate": 1.5700729927007297e-05,
|
| 937 |
+
"loss": 0.0337,
|
| 938 |
+
"step": 5200
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 1.4149659863945578,
|
| 942 |
+
"eval_loss": 0.030900483950972557,
|
| 943 |
+
"eval_runtime": 10.2164,
|
| 944 |
+
"eval_samples_per_second": 58.729,
|
| 945 |
+
"eval_steps_per_second": 7.341,
|
| 946 |
+
"step": 5200
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.4285714285714286,
|
| 950 |
+
"grad_norm": 0.3276965916156769,
|
| 951 |
+
"learning_rate": 1.5335766423357665e-05,
|
| 952 |
+
"loss": 0.0347,
|
| 953 |
+
"step": 5250
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 1.4421768707482994,
|
| 957 |
+
"grad_norm": 0.421546071767807,
|
| 958 |
+
"learning_rate": 1.497080291970803e-05,
|
| 959 |
+
"loss": 0.0328,
|
| 960 |
+
"step": 5300
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.4557823129251701,
|
| 964 |
+
"grad_norm": 0.33375367522239685,
|
| 965 |
+
"learning_rate": 1.4605839416058394e-05,
|
| 966 |
+
"loss": 0.0338,
|
| 967 |
+
"step": 5350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 1.469387755102041,
|
| 971 |
+
"grad_norm": 0.28487250208854675,
|
| 972 |
+
"learning_rate": 1.424087591240876e-05,
|
| 973 |
+
"loss": 0.0338,
|
| 974 |
+
"step": 5400
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.469387755102041,
|
| 978 |
+
"eval_loss": 0.030995788052678108,
|
| 979 |
+
"eval_runtime": 10.198,
|
| 980 |
+
"eval_samples_per_second": 58.835,
|
| 981 |
+
"eval_steps_per_second": 7.354,
|
| 982 |
+
"step": 5400
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.4829931972789114,
|
| 986 |
+
"grad_norm": 0.4164125919342041,
|
| 987 |
+
"learning_rate": 1.3875912408759125e-05,
|
| 988 |
+
"loss": 0.0347,
|
| 989 |
+
"step": 5450
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.4965986394557822,
|
| 993 |
+
"grad_norm": 0.5844776630401611,
|
| 994 |
+
"learning_rate": 1.3510948905109488e-05,
|
| 995 |
+
"loss": 0.0342,
|
| 996 |
+
"step": 5500
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.510204081632653,
|
| 1000 |
+
"grad_norm": 0.4449387490749359,
|
| 1001 |
+
"learning_rate": 1.3145985401459854e-05,
|
| 1002 |
+
"loss": 0.0329,
|
| 1003 |
+
"step": 5550
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.5238095238095237,
|
| 1007 |
+
"grad_norm": 0.351698100566864,
|
| 1008 |
+
"learning_rate": 1.2781021897810219e-05,
|
| 1009 |
+
"loss": 0.0334,
|
| 1010 |
+
"step": 5600
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.5238095238095237,
|
| 1014 |
+
"eval_loss": 0.030806375667452812,
|
| 1015 |
+
"eval_runtime": 10.1839,
|
| 1016 |
+
"eval_samples_per_second": 58.917,
|
| 1017 |
+
"eval_steps_per_second": 7.365,
|
| 1018 |
+
"step": 5600
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.5374149659863945,
|
| 1022 |
+
"grad_norm": 0.32961511611938477,
|
| 1023 |
+
"learning_rate": 1.2416058394160585e-05,
|
| 1024 |
+
"loss": 0.0332,
|
| 1025 |
+
"step": 5650
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.5510204081632653,
|
| 1029 |
+
"grad_norm": 0.39127564430236816,
|
| 1030 |
+
"learning_rate": 1.205109489051095e-05,
|
| 1031 |
+
"loss": 0.0329,
|
| 1032 |
+
"step": 5700
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.564625850340136,
|
| 1036 |
+
"grad_norm": 0.47163116931915283,
|
| 1037 |
+
"learning_rate": 1.1686131386861314e-05,
|
| 1038 |
+
"loss": 0.0328,
|
| 1039 |
+
"step": 5750
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.5782312925170068,
|
| 1043 |
+
"grad_norm": 0.34902673959732056,
|
| 1044 |
+
"learning_rate": 1.132116788321168e-05,
|
| 1045 |
+
"loss": 0.0341,
|
| 1046 |
+
"step": 5800
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.5782312925170068,
|
| 1050 |
+
"eval_loss": 0.030648473650217056,
|
| 1051 |
+
"eval_runtime": 10.232,
|
| 1052 |
+
"eval_samples_per_second": 58.639,
|
| 1053 |
+
"eval_steps_per_second": 7.33,
|
| 1054 |
+
"step": 5800
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.5918367346938775,
|
| 1058 |
+
"grad_norm": 0.34722593426704407,
|
| 1059 |
+
"learning_rate": 1.0956204379562045e-05,
|
| 1060 |
+
"loss": 0.0336,
|
| 1061 |
+
"step": 5850
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.6054421768707483,
|
| 1065 |
+
"grad_norm": 0.3117406964302063,
|
| 1066 |
+
"learning_rate": 1.0591240875912409e-05,
|
| 1067 |
+
"loss": 0.0325,
|
| 1068 |
+
"step": 5900
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.619047619047619,
|
| 1072 |
+
"grad_norm": 0.38354817032814026,
|
| 1073 |
+
"learning_rate": 1.0226277372262774e-05,
|
| 1074 |
+
"loss": 0.0337,
|
| 1075 |
+
"step": 5950
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.6326530612244898,
|
| 1079 |
+
"grad_norm": 0.4977235794067383,
|
| 1080 |
+
"learning_rate": 9.86131386861314e-06,
|
| 1081 |
+
"loss": 0.033,
|
| 1082 |
+
"step": 6000
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.6326530612244898,
|
| 1086 |
+
"eval_loss": 0.03188992664217949,
|
| 1087 |
+
"eval_runtime": 10.1863,
|
| 1088 |
+
"eval_samples_per_second": 58.903,
|
| 1089 |
+
"eval_steps_per_second": 7.363,
|
| 1090 |
+
"step": 6000
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.6462585034013606,
|
| 1094 |
+
"grad_norm": 0.3820977509021759,
|
| 1095 |
+
"learning_rate": 9.496350364963503e-06,
|
| 1096 |
+
"loss": 0.034,
|
| 1097 |
+
"step": 6050
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.6598639455782314,
|
| 1101 |
+
"grad_norm": 0.32841384410858154,
|
| 1102 |
+
"learning_rate": 9.131386861313869e-06,
|
| 1103 |
+
"loss": 0.0332,
|
| 1104 |
+
"step": 6100
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.6734693877551021,
|
| 1108 |
+
"grad_norm": 0.31902793049812317,
|
| 1109 |
+
"learning_rate": 8.766423357664235e-06,
|
| 1110 |
+
"loss": 0.0327,
|
| 1111 |
+
"step": 6150
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.6870748299319729,
|
| 1115 |
+
"grad_norm": 0.35828185081481934,
|
| 1116 |
+
"learning_rate": 8.401459854014598e-06,
|
| 1117 |
+
"loss": 0.0335,
|
| 1118 |
+
"step": 6200
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.6870748299319729,
|
| 1122 |
+
"eval_loss": 0.03037342056632042,
|
| 1123 |
+
"eval_runtime": 10.2013,
|
| 1124 |
+
"eval_samples_per_second": 58.816,
|
| 1125 |
+
"eval_steps_per_second": 7.352,
|
| 1126 |
+
"step": 6200
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 1.7006802721088436,
|
| 1130 |
+
"grad_norm": 0.3647211790084839,
|
| 1131 |
+
"learning_rate": 8.036496350364964e-06,
|
| 1132 |
+
"loss": 0.0331,
|
| 1133 |
+
"step": 6250
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 1.7142857142857144,
|
| 1137 |
+
"grad_norm": 0.34833037853240967,
|
| 1138 |
+
"learning_rate": 7.671532846715329e-06,
|
| 1139 |
+
"loss": 0.0344,
|
| 1140 |
+
"step": 6300
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.7278911564625852,
|
| 1144 |
+
"grad_norm": 0.35880446434020996,
|
| 1145 |
+
"learning_rate": 7.306569343065693e-06,
|
| 1146 |
+
"loss": 0.0326,
|
| 1147 |
+
"step": 6350
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 1.741496598639456,
|
| 1151 |
+
"grad_norm": 0.31057825684547424,
|
| 1152 |
+
"learning_rate": 6.941605839416059e-06,
|
| 1153 |
+
"loss": 0.0322,
|
| 1154 |
+
"step": 6400
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.741496598639456,
|
| 1158 |
+
"eval_loss": 0.030219364911317825,
|
| 1159 |
+
"eval_runtime": 10.2032,
|
| 1160 |
+
"eval_samples_per_second": 58.805,
|
| 1161 |
+
"eval_steps_per_second": 7.351,
|
| 1162 |
+
"step": 6400
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 1.7551020408163265,
|
| 1166 |
+
"grad_norm": 0.3011321723461151,
|
| 1167 |
+
"learning_rate": 6.576642335766424e-06,
|
| 1168 |
+
"loss": 0.0325,
|
| 1169 |
+
"step": 6450
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 1.7687074829931972,
|
| 1173 |
+
"grad_norm": 0.26797881722450256,
|
| 1174 |
+
"learning_rate": 6.2116788321167885e-06,
|
| 1175 |
+
"loss": 0.0322,
|
| 1176 |
+
"step": 6500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 1.782312925170068,
|
| 1180 |
+
"grad_norm": 0.4549163579940796,
|
| 1181 |
+
"learning_rate": 5.846715328467153e-06,
|
| 1182 |
+
"loss": 0.0325,
|
| 1183 |
+
"step": 6550
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 1.7959183673469388,
|
| 1187 |
+
"grad_norm": 0.3827670216560364,
|
| 1188 |
+
"learning_rate": 5.481751824817518e-06,
|
| 1189 |
+
"loss": 0.0329,
|
| 1190 |
+
"step": 6600
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 1.7959183673469388,
|
| 1194 |
+
"eval_loss": 0.030094588175415993,
|
| 1195 |
+
"eval_runtime": 10.1718,
|
| 1196 |
+
"eval_samples_per_second": 58.987,
|
| 1197 |
+
"eval_steps_per_second": 7.373,
|
| 1198 |
+
"step": 6600
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 1.8095238095238095,
|
| 1202 |
+
"grad_norm": 0.7711276412010193,
|
| 1203 |
+
"learning_rate": 5.116788321167884e-06,
|
| 1204 |
+
"loss": 0.0327,
|
| 1205 |
+
"step": 6650
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 1.8231292517006803,
|
| 1209 |
+
"grad_norm": 0.4340929388999939,
|
| 1210 |
+
"learning_rate": 4.751824817518248e-06,
|
| 1211 |
+
"loss": 0.0326,
|
| 1212 |
+
"step": 6700
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 1.836734693877551,
|
| 1216 |
+
"grad_norm": 0.46433642506599426,
|
| 1217 |
+
"learning_rate": 4.386861313868614e-06,
|
| 1218 |
+
"loss": 0.0331,
|
| 1219 |
+
"step": 6750
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.8503401360544216,
|
| 1223 |
+
"grad_norm": 0.5542903542518616,
|
| 1224 |
+
"learning_rate": 4.021897810218978e-06,
|
| 1225 |
+
"loss": 0.0336,
|
| 1226 |
+
"step": 6800
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 1.8503401360544216,
|
| 1230 |
+
"eval_loss": 0.029951954260468483,
|
| 1231 |
+
"eval_runtime": 10.1915,
|
| 1232 |
+
"eval_samples_per_second": 58.872,
|
| 1233 |
+
"eval_steps_per_second": 7.359,
|
| 1234 |
+
"step": 6800
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.8639455782312924,
|
| 1238 |
+
"grad_norm": 0.3692566156387329,
|
| 1239 |
+
"learning_rate": 3.6569343065693436e-06,
|
| 1240 |
+
"loss": 0.0322,
|
| 1241 |
+
"step": 6850
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.8775510204081631,
|
| 1245 |
+
"grad_norm": 0.25918880105018616,
|
| 1246 |
+
"learning_rate": 3.291970802919708e-06,
|
| 1247 |
+
"loss": 0.0325,
|
| 1248 |
+
"step": 6900
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.891156462585034,
|
| 1252 |
+
"grad_norm": 0.32452937960624695,
|
| 1253 |
+
"learning_rate": 2.927007299270073e-06,
|
| 1254 |
+
"loss": 0.0321,
|
| 1255 |
+
"step": 6950
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.9047619047619047,
|
| 1259 |
+
"grad_norm": 0.3027072250843048,
|
| 1260 |
+
"learning_rate": 2.562043795620438e-06,
|
| 1261 |
+
"loss": 0.0328,
|
| 1262 |
+
"step": 7000
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.9047619047619047,
|
| 1266 |
+
"eval_loss": 0.029974693432450294,
|
| 1267 |
+
"eval_runtime": 10.1955,
|
| 1268 |
+
"eval_samples_per_second": 58.849,
|
| 1269 |
+
"eval_steps_per_second": 7.356,
|
| 1270 |
+
"step": 7000
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 50,
|
| 1274 |
+
"max_steps": 7350,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 2,
|
| 1277 |
+
"save_steps": 200,
|
| 1278 |
+
"stateful_callbacks": {
|
| 1279 |
+
"TrainerControl": {
|
| 1280 |
+
"args": {
|
| 1281 |
+
"should_epoch_stop": false,
|
| 1282 |
+
"should_evaluate": false,
|
| 1283 |
+
"should_log": false,
|
| 1284 |
+
"should_save": true,
|
| 1285 |
+
"should_training_stop": false
|
| 1286 |
+
},
|
| 1287 |
+
"attributes": {}
|
| 1288 |
+
}
|
| 1289 |
+
},
|
| 1290 |
+
"total_flos": 2.9264707584e+16,
|
| 1291 |
+
"train_batch_size": 2,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|
checkpoints/checkpoint-step-7000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e65f11e00edb82d09a337df334551a2d5eac2eb0f7f94aaa44d3be5e86cc7a7
|
| 3 |
+
size 5777
|
checkpoints/checkpoint-step-7000/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|