Training in progress, step 9000, checkpoint

checkpoint-9000/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-9000/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-9000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b86ec7abf140461faca889d333b6c3ee51deb3192c2ff959957ab75ab9e0d3ba
+size 435544704

checkpoint-9000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61115f1e8513f113d894252940c1898147ec28920d72e2ba613709fe7392f18c
+size 871183627

checkpoint-9000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e22020f6ba1e9a309a7d5c1e38d9f1cb1cea3c2b6ce080ebd400c8f7dc88b94
+size 14645

checkpoint-9000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea10f4b4a5ea7ad7fae1aaad10310297ce6dc976c7952f0f8a2bc62de2d8f322
+size 1383

checkpoint-9000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e87bfc9eadd2c880b7a709220090aa63017c9c27cf77d83d1763f63175ebe98
+size 1465

checkpoint-9000/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-9000/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-9000/trainer_state.json

@@ -0,0 +1,311 @@
+{
+  "best_global_step": 4464,
+  "best_metric": 3.094606399536133,
+  "best_model_checkpoint": null,
+  "epoch": 18.14516129032258,
+  "eval_steps": 500,
+  "global_step": 9000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0020161290322580645,
+      "grad_norm": 28.973712921142578,
+      "learning_rate": 0.0,
+      "loss": 10.4319,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 5.415546417236328,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 4.585,
+      "step": 496
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.5159876346588135,
+      "eval_runtime": 19.7164,
+      "eval_samples_per_second": 1614.748,
+      "eval_steps_per_second": 6.34,
+      "step": 496
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.859699249267578,
+      "learning_rate": 9.477659574468086e-05,
+      "loss": 3.3235,
+      "step": 992
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.300523519515991,
+      "eval_runtime": 19.7153,
+      "eval_samples_per_second": 1614.835,
+      "eval_steps_per_second": 6.34,
+      "step": 992
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 4.334104061126709,
+      "learning_rate": 8.950000000000001e-05,
+      "loss": 3.1692,
+      "step": 1488
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.223573923110962,
+      "eval_runtime": 19.6395,
+      "eval_samples_per_second": 1621.068,
+      "eval_steps_per_second": 6.365,
+      "step": 1488
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.5572965145111084,
+      "learning_rate": 8.422340425531914e-05,
+      "loss": 3.0793,
+      "step": 1984
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.17850399017334,
+      "eval_runtime": 19.9743,
+      "eval_samples_per_second": 1593.897,
+      "eval_steps_per_second": 6.258,
+      "step": 1984
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 4.0843071937561035,
+      "learning_rate": 7.894680851063831e-05,
+      "loss": 3.0144,
+      "step": 2480
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.1462278366088867,
+      "eval_runtime": 19.75,
+      "eval_samples_per_second": 1611.996,
+      "eval_steps_per_second": 6.329,
+      "step": 2480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.260258197784424,
+      "learning_rate": 7.367021276595744e-05,
+      "loss": 2.9616,
+      "step": 2976
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.121847152709961,
+      "eval_runtime": 19.7368,
+      "eval_samples_per_second": 1613.075,
+      "eval_steps_per_second": 6.333,
+      "step": 2976
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.405400276184082,
+      "learning_rate": 6.83936170212766e-05,
+      "loss": 2.9137,
+      "step": 3472
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.1083970069885254,
+      "eval_runtime": 19.7383,
+      "eval_samples_per_second": 1612.955,
+      "eval_steps_per_second": 6.333,
+      "step": 3472
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.643365859985352,
+      "learning_rate": 6.311702127659575e-05,
+      "loss": 2.8691,
+      "step": 3968
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.099594831466675,
+      "eval_runtime": 19.944,
+      "eval_samples_per_second": 1596.318,
+      "eval_steps_per_second": 6.268,
+      "step": 3968
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.5937089920043945,
+      "learning_rate": 5.7840425531914896e-05,
+      "loss": 2.8269,
+      "step": 4464
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.094606399536133,
+      "eval_runtime": 19.8149,
+      "eval_samples_per_second": 1606.72,
+      "eval_steps_per_second": 6.308,
+      "step": 4464
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 4.907343864440918,
+      "learning_rate": 5.256382978723404e-05,
+      "loss": 2.7864,
+      "step": 4960
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.096388578414917,
+      "eval_runtime": 19.755,
+      "eval_samples_per_second": 1611.588,
+      "eval_steps_per_second": 6.327,
+      "step": 4960
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 5.795402526855469,
+      "learning_rate": 4.728723404255319e-05,
+      "loss": 2.7482,
+      "step": 5456
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.1013050079345703,
+      "eval_runtime": 19.7774,
+      "eval_samples_per_second": 1609.768,
+      "eval_steps_per_second": 6.32,
+      "step": 5456
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 6.2599101066589355,
+      "learning_rate": 4.201063829787234e-05,
+      "loss": 2.7091,
+      "step": 5952
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.1037847995758057,
+      "eval_runtime": 19.7562,
+      "eval_samples_per_second": 1611.496,
+      "eval_steps_per_second": 6.327,
+      "step": 5952
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 5.863295078277588,
+      "learning_rate": 3.673404255319149e-05,
+      "loss": 2.6712,
+      "step": 6448
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.1146838665008545,
+      "eval_runtime": 19.7597,
+      "eval_samples_per_second": 1611.21,
+      "eval_steps_per_second": 6.326,
+      "step": 6448
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 7.292373180389404,
+      "learning_rate": 3.1457446808510635e-05,
+      "loss": 2.637,
+      "step": 6944
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.1215128898620605,
+      "eval_runtime": 19.8049,
+      "eval_samples_per_second": 1607.533,
+      "eval_steps_per_second": 6.312,
+      "step": 6944
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 8.47721004486084,
+      "learning_rate": 2.6180851063829785e-05,
+      "loss": 2.6037,
+      "step": 7440
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.1334593296051025,
+      "eval_runtime": 19.752,
+      "eval_samples_per_second": 1611.836,
+      "eval_steps_per_second": 6.328,
+      "step": 7440
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 7.659241676330566,
+      "learning_rate": 2.090425531914894e-05,
+      "loss": 2.571,
+      "step": 7936
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.1450557708740234,
+      "eval_runtime": 19.7737,
+      "eval_samples_per_second": 1610.07,
+      "eval_steps_per_second": 6.322,
+      "step": 7936
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": Infinity,
+      "learning_rate": 1.5627659574468086e-05,
+      "loss": 2.5432,
+      "step": 8432
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.1547369956970215,
+      "eval_runtime": 19.7694,
+      "eval_samples_per_second": 1610.417,
+      "eval_steps_per_second": 6.323,
+      "step": 8432
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 8.386995315551758,
+      "learning_rate": 1.0361702127659576e-05,
+      "loss": 2.5189,
+      "step": 8928
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 3.165121555328369,
+      "eval_runtime": 19.7554,
+      "eval_samples_per_second": 1611.557,
+      "eval_steps_per_second": 6.327,
+      "step": 8928
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9920,
+  "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": 1.50207905267712e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-9000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d07957fe339e96ebd2ea946f523ff0d3feb5cdd484ea9f5d836fa3d8b77a54df
+size 5905