init files

README.md

@@ -1,3 +1,58 @@
 ---
-license: mit
+tags:
+- generated_from_trainer
+datasets:
+- RaiBP/openwebtext2-first-30-chunks-ablation-bilingual
+model-index:
+- name: training_bilingual
+  results: []
 ---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# training_bilingual
+
+This model was trained from scratch on the RaiBP/openwebtext2-first-30-chunks-ablation-bilingual dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.005
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 1000
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 1.13.0
+- Datasets 2.16.0
+- Tokenizers 0.15.0

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 3.3394907157594433,
+    "train_runtime": 176655.7796,
+    "train_samples": 2963317,
+    "train_samples_per_second": 16.775,
+    "train_steps_per_second": 0.262
+}

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "./training",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.0,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.0,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 50257
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.37.0.dev0"
+}

runs/events.out.tfevents.1704797850.pytorchRB.15471.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c89a3935f7280a62913d7a35f8bd8a079dcc2746c68c95c5d2c6b4d453232345
+size 19590

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,20 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 3.3394907157594433,
+    "train_runtime": 176655.7796,
+    "train_samples": 2963317,
+    "train_samples_per_second": 16.775,
+    "train_steps_per_second": 0.262
+}

trainer_state.json

@@ -0,0 +1,582 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 46302,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.0025,
+      "loss": 6.1558,
+      "step": 500
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.005,
+      "loss": 4.891,
+      "step": 1000
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.004944814798463645,
+      "loss": 4.3187,
+      "step": 1500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.004889629596927288,
+      "loss": 4.0557,
+      "step": 2000
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.004834444395390932,
+      "loss": 3.9101,
+      "step": 2500
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.004779259193854576,
+      "loss": 3.8112,
+      "step": 3000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00472407399231822,
+      "loss": 3.7468,
+      "step": 3500
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.004668888790781864,
+      "loss": 3.6894,
+      "step": 4000
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0046137035892455085,
+      "loss": 3.6476,
+      "step": 4500
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.004558518387709152,
+      "loss": 3.6159,
+      "step": 5000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.004503333186172796,
+      "loss": 3.5868,
+      "step": 5500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00444814798463644,
+      "loss": 3.5591,
+      "step": 6000
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.004392962783100084,
+      "loss": 3.5347,
+      "step": 6500
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.004337777581563728,
+      "loss": 3.515,
+      "step": 7000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0042825923800273715,
+      "loss": 3.4965,
+      "step": 7500
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.004227407178491016,
+      "loss": 3.4824,
+      "step": 8000
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00417222197695466,
+      "loss": 3.4681,
+      "step": 8500
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.004117036775418304,
+      "loss": 3.4506,
+      "step": 9000
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.004061851573881948,
+      "loss": 3.4438,
+      "step": 9500
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.004006666372345592,
+      "loss": 3.4277,
+      "step": 10000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.003951481170809235,
+      "loss": 3.4096,
+      "step": 10500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00389629596927288,
+      "loss": 3.4064,
+      "step": 11000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0038411107677365237,
+      "loss": 3.3917,
+      "step": 11500
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.003785925566200168,
+      "loss": 3.3817,
+      "step": 12000
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0037307403646638115,
+      "loss": 3.3771,
+      "step": 12500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.003675555163127456,
+      "loss": 3.3711,
+      "step": 13000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0036203699615910997,
+      "loss": 3.3564,
+      "step": 13500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.003565184760054744,
+      "loss": 3.3457,
+      "step": 14000
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.003509999558518388,
+      "loss": 3.3417,
+      "step": 14500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0034548143569820317,
+      "loss": 3.3381,
+      "step": 15000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.003399629155445676,
+      "loss": 3.3267,
+      "step": 15500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0033444439539093195,
+      "loss": 3.3211,
+      "step": 16000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.003289258752372964,
+      "loss": 3.3143,
+      "step": 16500
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0032340735508366077,
+      "loss": 3.3111,
+      "step": 17000
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.003178888349300252,
+      "loss": 3.3028,
+      "step": 17500
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0031237031477638955,
+      "loss": 3.2986,
+      "step": 18000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0030685179462275397,
+      "loss": 3.2882,
+      "step": 18500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0030133327446911834,
+      "loss": 3.2841,
+      "step": 19000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.002958147543154828,
+      "loss": 3.2856,
+      "step": 19500
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.002902962341618471,
+      "loss": 3.2793,
+      "step": 20000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0028477771400821157,
+      "loss": 3.2692,
+      "step": 20500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00279259193854576,
+      "loss": 3.2682,
+      "step": 21000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0027374067370094035,
+      "loss": 3.2577,
+      "step": 21500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0026822215354730477,
+      "loss": 3.2555,
+      "step": 22000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0026270363339366914,
+      "loss": 3.2485,
+      "step": 22500
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.002571851132400336,
+      "loss": 3.2393,
+      "step": 23000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0025166659308639796,
+      "loss": 3.2336,
+      "step": 23500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0024614807293276237,
+      "loss": 3.2298,
+      "step": 24000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.002406295527791268,
+      "loss": 3.2306,
+      "step": 24500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0023511103262549115,
+      "loss": 3.2263,
+      "step": 25000
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.0022959251247185557,
+      "loss": 3.2168,
+      "step": 25500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0022407399231821994,
+      "loss": 3.2132,
+      "step": 26000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0021855547216458435,
+      "loss": 3.2078,
+      "step": 26500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0021303695201094876,
+      "loss": 3.2056,
+      "step": 27000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0020751843185731313,
+      "loss": 3.2043,
+      "step": 27500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0020199991170367754,
+      "loss": 3.1978,
+      "step": 28000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0019648139155004195,
+      "loss": 3.1933,
+      "step": 28500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0019096287139640635,
+      "loss": 3.1942,
+      "step": 29000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0018544435124277074,
+      "loss": 3.1849,
+      "step": 29500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0017992583108913513,
+      "loss": 3.1805,
+      "step": 30000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0017440731093549956,
+      "loss": 3.176,
+      "step": 30500
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0016888879078186395,
+      "loss": 3.1668,
+      "step": 31000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0016337027062822834,
+      "loss": 3.1643,
+      "step": 31500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0015785175047459273,
+      "loss": 3.1624,
+      "step": 32000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0015233323032095715,
+      "loss": 3.1581,
+      "step": 32500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0014681471016732154,
+      "loss": 3.15,
+      "step": 33000
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0014129619001368593,
+      "loss": 3.1471,
+      "step": 33500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0013577766986005034,
+      "loss": 3.1483,
+      "step": 34000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0013025914970641473,
+      "loss": 3.1404,
+      "step": 34500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0012474062955277914,
+      "loss": 3.1312,
+      "step": 35000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0011922210939914353,
+      "loss": 3.133,
+      "step": 35500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0011370358924550792,
+      "loss": 3.1267,
+      "step": 36000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0010818506909187234,
+      "loss": 3.1242,
+      "step": 36500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0010266654893823673,
+      "loss": 3.1176,
+      "step": 37000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0009714802878460112,
+      "loss": 3.1104,
+      "step": 37500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0009162950863096552,
+      "loss": 3.1077,
+      "step": 38000
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0008611098847732993,
+      "loss": 3.1033,
+      "step": 38500
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0008059246832369432,
+      "loss": 3.0973,
+      "step": 39000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0007507394817005872,
+      "loss": 3.0969,
+      "step": 39500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0006955542801642311,
+      "loss": 3.0919,
+      "step": 40000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0006403690786278752,
+      "loss": 3.0814,
+      "step": 40500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0005851838770915192,
+      "loss": 3.0843,
+      "step": 41000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0005299986755551632,
+      "loss": 3.0756,
+      "step": 41500
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0004748134740188071,
+      "loss": 3.0765,
+      "step": 42000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0004196282724824511,
+      "loss": 3.0672,
+      "step": 42500
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00036444307094609513,
+      "loss": 3.0666,
+      "step": 43000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0003092578694097391,
+      "loss": 3.0646,
+      "step": 43500
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00025407266787338305,
+      "loss": 3.0544,
+      "step": 44000
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00019888746633702707,
+      "loss": 3.0552,
+      "step": 44500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00014370226480067106,
+      "loss": 3.0548,
+      "step": 45000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 8.851706326431504e-05,
+      "loss": 3.0448,
+      "step": 45500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 3.333186172795903e-05,
+      "loss": 3.0407,
+      "step": 46000
+    },
+    {
+      "epoch": 1.0,
+      "step": 46302,
+      "total_flos": 1.5485880137668362e+18,
+      "train_loss": 3.3394907157594433,
+      "train_runtime": 176655.7796,
+      "train_samples_per_second": 16.775,
+      "train_steps_per_second": 0.262
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 46302,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 5000,
+  "total_flos": 1.5485880137668362e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8074368a0d7d5119c9fed40f9eb92e0fd7985288ebad508a595eafcd45f57400
+size 4219