Training in progress, step 7980, checkpoint

checkpoint-7980/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-7980/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-7980/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:174a0b088e4d1679e78e162879fc78d21289e4d5656bf5c2d4c1bca729086ec4
+size 435544704

checkpoint-7980/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e822faadf8515d54143d1229f935b961aa9584b27800429169a2f996a394e2
+size 871183627

checkpoint-7980/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bae7ce73c2cf603849db50b674f6744b93a1f40ec80372785f0b95c4aaed33e3
+size 14645

checkpoint-7980/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd66dab9e9319af067434178d43a0f65cb09e43fb4159f93f4611d07cb83f8b4
+size 1383

checkpoint-7980/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63bf0b640c93a96c51c4d80210bc0946fee83c495c5efd8b2af11c6dab457105
+size 1465

checkpoint-7980/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-7980/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-7980/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 5187,
+  "best_metric": 5.9927754402160645,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 7980,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002506265664160401,
+      "grad_norm": 15.720861434936523,
+      "learning_rate": 0.0,
+      "loss": 10.3873,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.5908706188201904,
+      "learning_rate": 7.960000000000001e-05,
+      "loss": 7.3511,
+      "step": 399
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.7809553146362305,
+      "eval_runtime": 15.7948,
+      "eval_samples_per_second": 1618.759,
+      "eval_steps_per_second": 6.331,
+      "step": 399
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.6086559295654297,
+      "learning_rate": 9.601876675603218e-05,
+      "loss": 6.5625,
+      "step": 798
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 6.501540184020996,
+      "eval_runtime": 15.7225,
+      "eval_samples_per_second": 1626.203,
+      "eval_steps_per_second": 6.36,
+      "step": 798
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.726973295211792,
+      "learning_rate": 9.067024128686328e-05,
+      "loss": 6.3574,
+      "step": 1197
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 6.369043350219727,
+      "eval_runtime": 15.8045,
+      "eval_samples_per_second": 1617.763,
+      "eval_steps_per_second": 6.327,
+      "step": 1197
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9310587644577026,
+      "learning_rate": 8.532171581769438e-05,
+      "loss": 6.2341,
+      "step": 1596
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 6.280962944030762,
+      "eval_runtime": 15.8456,
+      "eval_samples_per_second": 1613.567,
+      "eval_steps_per_second": 6.311,
+      "step": 1596
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.955281376838684,
+      "learning_rate": 7.997319034852548e-05,
+      "loss": 6.1366,
+      "step": 1995
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 6.2110114097595215,
+      "eval_runtime": 15.7542,
+      "eval_samples_per_second": 1622.935,
+      "eval_steps_per_second": 6.348,
+      "step": 1995
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.114393711090088,
+      "learning_rate": 7.462466487935658e-05,
+      "loss": 6.0506,
+      "step": 2394
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 6.1528520584106445,
+      "eval_runtime": 15.8299,
+      "eval_samples_per_second": 1615.167,
+      "eval_steps_per_second": 6.317,
+      "step": 2394
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.2936296463012695,
+      "learning_rate": 6.927613941018766e-05,
+      "loss": 5.9708,
+      "step": 2793
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 6.105339050292969,
+      "eval_runtime": 15.8332,
+      "eval_samples_per_second": 1614.839,
+      "eval_steps_per_second": 6.316,
+      "step": 2793
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.3843250274658203,
+      "learning_rate": 6.392761394101878e-05,
+      "loss": 5.896,
+      "step": 3192
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 6.064266681671143,
+      "eval_runtime": 15.7947,
+      "eval_samples_per_second": 1618.769,
+      "eval_steps_per_second": 6.331,
+      "step": 3192
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.7147932052612305,
+      "learning_rate": 5.8579088471849864e-05,
+      "loss": 5.8247,
+      "step": 3591
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 6.038033485412598,
+      "eval_runtime": 15.8518,
+      "eval_samples_per_second": 1612.942,
+      "eval_steps_per_second": 6.308,
+      "step": 3591
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 2.932281970977783,
+      "learning_rate": 5.3230563002680965e-05,
+      "loss": 5.7573,
+      "step": 3990
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 6.01782751083374,
+      "eval_runtime": 15.8517,
+      "eval_samples_per_second": 1612.948,
+      "eval_steps_per_second": 6.308,
+      "step": 3990
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 3.332143545150757,
+      "learning_rate": 4.7882037533512065e-05,
+      "loss": 5.6939,
+      "step": 4389
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 6.003720283508301,
+      "eval_runtime": 15.8585,
+      "eval_samples_per_second": 1612.263,
+      "eval_steps_per_second": 6.306,
+      "step": 4389
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 3.4687414169311523,
+      "learning_rate": 4.2533512064343165e-05,
+      "loss": 5.6329,
+      "step": 4788
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.996026515960693,
+      "eval_runtime": 15.8796,
+      "eval_samples_per_second": 1610.12,
+      "eval_steps_per_second": 6.297,
+      "step": 4788
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 3.8036763668060303,
+      "learning_rate": 3.7184986595174266e-05,
+      "loss": 5.5769,
+      "step": 5187
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.9927754402160645,
+      "eval_runtime": 19.177,
+      "eval_samples_per_second": 1333.262,
+      "eval_steps_per_second": 5.215,
+      "step": 5187
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 4.060362339019775,
+      "learning_rate": 3.1836461126005366e-05,
+      "loss": 5.5238,
+      "step": 5586
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.993233680725098,
+      "eval_runtime": 20.5877,
+      "eval_samples_per_second": 1241.904,
+      "eval_steps_per_second": 4.857,
+      "step": 5586
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 4.1933159828186035,
+      "learning_rate": 2.648793565683646e-05,
+      "loss": 5.4758,
+      "step": 5985
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.997311592102051,
+      "eval_runtime": 20.9055,
+      "eval_samples_per_second": 1223.028,
+      "eval_steps_per_second": 4.783,
+      "step": 5985
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 4.521027565002441,
+      "learning_rate": 2.113941018766756e-05,
+      "loss": 5.4308,
+      "step": 6384
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 6.001664638519287,
+      "eval_runtime": 20.6766,
+      "eval_samples_per_second": 1236.566,
+      "eval_steps_per_second": 4.836,
+      "step": 6384
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 4.66829776763916,
+      "learning_rate": 1.579088471849866e-05,
+      "loss": 5.3926,
+      "step": 6783
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 6.008192539215088,
+      "eval_runtime": 21.0082,
+      "eval_samples_per_second": 1217.049,
+      "eval_steps_per_second": 4.76,
+      "step": 6783
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 5.069458961486816,
+      "learning_rate": 1.044235924932976e-05,
+      "loss": 5.361,
+      "step": 7182
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 6.012160778045654,
+      "eval_runtime": 20.2794,
+      "eval_samples_per_second": 1260.786,
+      "eval_steps_per_second": 4.931,
+      "step": 7182
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 5.077199459075928,
+      "learning_rate": 5.093833780160858e-06,
+      "loss": 5.335,
+      "step": 7581
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 6.016257286071777,
+      "eval_runtime": 20.5337,
+      "eval_samples_per_second": 1245.174,
+      "eval_steps_per_second": 4.87,
+      "step": 7581
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7980,
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.33128290304e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7980/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78833db4bfee51bda1c0924b513a80cdd80ef3da9a024a85aed55bff9f66acbf
+size 5969