Training in progress, step 8000, checkpoint

checkpoint-8000/config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "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,
+  "pad_token_id": 2,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "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,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 30000
+}

checkpoint-8000/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "pad_token_id": 2,
+  "transformers_version": "4.56.1"
+}

checkpoint-8000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85ae074ecaf0b8d9b4817f9fd8a8ac89975d0628b05d07487841f1105fa15161
+size 435544704

checkpoint-8000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5965797ccd2bc642788b2f893d6c600930fc125d2ba6a33c79abf96e95ae68ba
+size 871183627

checkpoint-8000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ceba2154d2d3291409ed721383b6dda984b277476e489dc38cf332529226b48
+size 14645

checkpoint-8000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:022915fa8ac65fae0f13853519b8aa7ee79efd632a399c1f64c6b87bab0b057a
+size 1383

checkpoint-8000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7f8804ba5081440adcff1a5d9e3b413422dd06718c1d771d12f5877ebc3a329
+size 1465

checkpoint-8000/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

checkpoint-8000/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-8000/trainer_state.json

@@ -0,0 +1,281 @@
+{
+  "best_global_step": 5445,
+  "best_metric": 3.110133647918701,
+  "best_model_checkpoint": null,
+  "epoch": 16.161616161616163,
+  "eval_steps": 500,
+  "global_step": 8000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00202020202020202,
+      "grad_norm": 29.33669662475586,
+      "learning_rate": 0.0,
+      "loss": 10.325,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 3.660304546356201,
+      "learning_rate": 9.88e-05,
+      "loss": 4.902,
+      "step": 495
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.9300971031188965,
+      "eval_runtime": 30.1559,
+      "eval_samples_per_second": 1053.392,
+      "eval_steps_per_second": 4.145,
+      "step": 495
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.1966376304626465,
+      "learning_rate": 9.478678038379531e-05,
+      "loss": 3.6496,
+      "step": 990
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.54586124420166,
+      "eval_runtime": 30.3022,
+      "eval_samples_per_second": 1048.308,
+      "eval_steps_per_second": 4.125,
+      "step": 990
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.2939534187316895,
+      "learning_rate": 8.950959488272922e-05,
+      "loss": 3.3759,
+      "step": 1485
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.3782424926757812,
+      "eval_runtime": 30.2562,
+      "eval_samples_per_second": 1049.901,
+      "eval_steps_per_second": 4.131,
+      "step": 1485
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.3274989128112793,
+      "learning_rate": 8.42324093816631e-05,
+      "loss": 3.2227,
+      "step": 1980
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.2882602214813232,
+      "eval_runtime": 30.3221,
+      "eval_samples_per_second": 1047.618,
+      "eval_steps_per_second": 4.122,
+      "step": 1980
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.4603676795959473,
+      "learning_rate": 7.895522388059702e-05,
+      "loss": 3.1161,
+      "step": 2475
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.225015640258789,
+      "eval_runtime": 30.114,
+      "eval_samples_per_second": 1054.857,
+      "eval_steps_per_second": 4.151,
+      "step": 2475
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.4157865047454834,
+      "learning_rate": 7.367803837953093e-05,
+      "loss": 3.0318,
+      "step": 2970
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.186840772628784,
+      "eval_runtime": 30.7198,
+      "eval_samples_per_second": 1034.055,
+      "eval_steps_per_second": 4.069,
+      "step": 2970
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.641024112701416,
+      "learning_rate": 6.840085287846483e-05,
+      "loss": 2.9616,
+      "step": 3465
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.153538227081299,
+      "eval_runtime": 30.8772,
+      "eval_samples_per_second": 1028.784,
+      "eval_steps_per_second": 4.048,
+      "step": 3465
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.041162490844727,
+      "learning_rate": 6.312366737739872e-05,
+      "loss": 2.9001,
+      "step": 3960
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.1385979652404785,
+      "eval_runtime": 30.3668,
+      "eval_samples_per_second": 1046.077,
+      "eval_steps_per_second": 4.116,
+      "step": 3960
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.891669750213623,
+      "learning_rate": 5.784648187633263e-05,
+      "loss": 2.8433,
+      "step": 4455
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.1241445541381836,
+      "eval_runtime": 30.3151,
+      "eval_samples_per_second": 1047.86,
+      "eval_steps_per_second": 4.123,
+      "step": 4455
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 4.186176300048828,
+      "learning_rate": 5.256929637526653e-05,
+      "loss": 2.7912,
+      "step": 4950
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.118201971054077,
+      "eval_runtime": 30.9231,
+      "eval_samples_per_second": 1027.257,
+      "eval_steps_per_second": 4.042,
+      "step": 4950
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 4.513829708099365,
+      "learning_rate": 4.7292110874200426e-05,
+      "loss": 2.743,
+      "step": 5445
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.110133647918701,
+      "eval_runtime": 30.6226,
+      "eval_samples_per_second": 1037.338,
+      "eval_steps_per_second": 4.082,
+      "step": 5445
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 4.549180507659912,
+      "learning_rate": 4.201492537313433e-05,
+      "loss": 2.6958,
+      "step": 5940
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.111478805541992,
+      "eval_runtime": 30.7794,
+      "eval_samples_per_second": 1032.054,
+      "eval_steps_per_second": 4.061,
+      "step": 5940
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 4.832803726196289,
+      "learning_rate": 3.673773987206823e-05,
+      "loss": 2.6514,
+      "step": 6435
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.1139676570892334,
+      "eval_runtime": 30.2187,
+      "eval_samples_per_second": 1051.203,
+      "eval_steps_per_second": 4.137,
+      "step": 6435
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 4.833005428314209,
+      "learning_rate": 3.1460554371002134e-05,
+      "loss": 2.6104,
+      "step": 6930
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.1268951892852783,
+      "eval_runtime": 30.4657,
+      "eval_samples_per_second": 1042.682,
+      "eval_steps_per_second": 4.103,
+      "step": 6930
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 5.585130214691162,
+      "learning_rate": 2.6183368869936037e-05,
+      "loss": 2.5736,
+      "step": 7425
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.1311564445495605,
+      "eval_runtime": 33.0861,
+      "eval_samples_per_second": 960.101,
+      "eval_steps_per_second": 3.778,
+      "step": 7425
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 6.103977203369141,
+      "learning_rate": 2.0906183368869936e-05,
+      "loss": 2.5413,
+      "step": 7920
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.1396145820617676,
+      "eval_runtime": 32.1833,
+      "eval_samples_per_second": 987.033,
+      "eval_steps_per_second": 3.884,
+      "step": 7920
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9900,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.33524409024512e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-8000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7431b996de020d5431709eef1059b595118c677c31200e5c43e119c969068e4f
+size 5905