Training in progress, step 9900, checkpoint

checkpoint-9900/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-9900/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-9900/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9646509544e1d5035a3a2e75a36e841a1c905fa6e19d5adc037a511861664906
+size 435544704

checkpoint-9900/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee04657e649c508e5287944ba13d44b370df5c33a9820f6d248c762d6d46a416
+size 871183627

checkpoint-9900/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60c2eaf1455dff7a5f36cce87905d6a3511e0913e5beeffd31e7886ad98f93e5
+size 14645

checkpoint-9900/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a5ddab22d8a1c52369c0f9d8d0f6489321ff2d9a19616bd19271f43030cbbc
+size 1383

checkpoint-9900/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8732b31a5837ff2411de273d74dc3972f94fe566013c6ddc3264f1791c73ba4
+size 1465

checkpoint-9900/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-9900/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-9900/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 5445,
+  "best_metric": 2.7069296836853027,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 9900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00202020202020202,
+      "grad_norm": 30.17180061340332,
+      "learning_rate": 0.0,
+      "loss": 10.3764,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.0347892045974731,
+      "learning_rate": 9.88e-05,
+      "loss": 4.4691,
+      "step": 495
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.2804644107818604,
+      "eval_runtime": 24.5717,
+      "eval_samples_per_second": 1295.026,
+      "eval_steps_per_second": 5.087,
+      "step": 495
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.7221413850784302,
+      "learning_rate": 9.478678038379531e-05,
+      "loss": 3.0377,
+      "step": 990
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.978177547454834,
+      "eval_runtime": 24.5949,
+      "eval_samples_per_second": 1293.802,
+      "eval_steps_per_second": 5.082,
+      "step": 990
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7161000967025757,
+      "learning_rate": 8.950959488272922e-05,
+      "loss": 2.8107,
+      "step": 1485
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.865334987640381,
+      "eval_runtime": 24.5479,
+      "eval_samples_per_second": 1296.284,
+      "eval_steps_per_second": 5.092,
+      "step": 1485
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.70469069480896,
+      "learning_rate": 8.42324093816631e-05,
+      "loss": 2.6911,
+      "step": 1980
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.80560040473938,
+      "eval_runtime": 24.8232,
+      "eval_samples_per_second": 1281.906,
+      "eval_steps_per_second": 5.036,
+      "step": 1980
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7370979189872742,
+      "learning_rate": 7.895522388059702e-05,
+      "loss": 2.6078,
+      "step": 2475
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.7674310207366943,
+      "eval_runtime": 24.5579,
+      "eval_samples_per_second": 1295.752,
+      "eval_steps_per_second": 5.09,
+      "step": 2475
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.7106032371520996,
+      "learning_rate": 7.367803837953093e-05,
+      "loss": 2.5419,
+      "step": 2970
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.7415506839752197,
+      "eval_runtime": 24.5157,
+      "eval_samples_per_second": 1297.986,
+      "eval_steps_per_second": 5.099,
+      "step": 2970
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.7335968613624573,
+      "learning_rate": 6.840085287846483e-05,
+      "loss": 2.4854,
+      "step": 3465
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.7231545448303223,
+      "eval_runtime": 24.5148,
+      "eval_samples_per_second": 1298.032,
+      "eval_steps_per_second": 5.099,
+      "step": 3465
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.7922199964523315,
+      "learning_rate": 6.312366737739872e-05,
+      "loss": 2.4346,
+      "step": 3960
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.7113661766052246,
+      "eval_runtime": 24.7154,
+      "eval_samples_per_second": 1287.496,
+      "eval_steps_per_second": 5.058,
+      "step": 3960
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.8236603736877441,
+      "learning_rate": 5.784648187633263e-05,
+      "loss": 2.3875,
+      "step": 4455
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.7084221839904785,
+      "eval_runtime": 24.5194,
+      "eval_samples_per_second": 1297.787,
+      "eval_steps_per_second": 5.098,
+      "step": 4455
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.8995434641838074,
+      "learning_rate": 5.256929637526653e-05,
+      "loss": 2.3426,
+      "step": 4950
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.7084527015686035,
+      "eval_runtime": 24.5477,
+      "eval_samples_per_second": 1296.292,
+      "eval_steps_per_second": 5.092,
+      "step": 4950
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.9311981797218323,
+      "learning_rate": 4.7292110874200426e-05,
+      "loss": 2.2995,
+      "step": 5445
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.7069296836853027,
+      "eval_runtime": 24.4861,
+      "eval_samples_per_second": 1299.556,
+      "eval_steps_per_second": 5.105,
+      "step": 5445
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 0.984067440032959,
+      "learning_rate": 4.201492537313433e-05,
+      "loss": 2.2582,
+      "step": 5940
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.7145166397094727,
+      "eval_runtime": 24.6529,
+      "eval_samples_per_second": 1290.76,
+      "eval_steps_per_second": 5.07,
+      "step": 5940
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.0923529863357544,
+      "learning_rate": 3.673773987206823e-05,
+      "loss": 2.219,
+      "step": 6435
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 2.7212188243865967,
+      "eval_runtime": 24.446,
+      "eval_samples_per_second": 1301.683,
+      "eval_steps_per_second": 5.113,
+      "step": 6435
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.1309531927108765,
+      "learning_rate": 3.1460554371002134e-05,
+      "loss": 2.1812,
+      "step": 6930
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 2.726654291152954,
+      "eval_runtime": 24.4433,
+      "eval_samples_per_second": 1301.829,
+      "eval_steps_per_second": 5.114,
+      "step": 6930
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.1805713176727295,
+      "learning_rate": 2.6183368869936037e-05,
+      "loss": 2.1452,
+      "step": 7425
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.741478443145752,
+      "eval_runtime": 24.5647,
+      "eval_samples_per_second": 1295.394,
+      "eval_steps_per_second": 5.089,
+      "step": 7425
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.2184388637542725,
+      "learning_rate": 2.0906183368869936e-05,
+      "loss": 2.1129,
+      "step": 7920
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 2.749194383621216,
+      "eval_runtime": 24.6922,
+      "eval_samples_per_second": 1288.709,
+      "eval_steps_per_second": 5.062,
+      "step": 7920
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.2942442893981934,
+      "learning_rate": 1.562899786780384e-05,
+      "loss": 2.0848,
+      "step": 8415
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 2.76253604888916,
+      "eval_runtime": 24.9926,
+      "eval_samples_per_second": 1273.219,
+      "eval_steps_per_second": 5.001,
+      "step": 8415
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.3517400026321411,
+      "learning_rate": 1.035181236673774e-05,
+      "loss": 2.0606,
+      "step": 8910
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 2.769240379333496,
+      "eval_runtime": 24.6248,
+      "eval_samples_per_second": 1292.236,
+      "eval_steps_per_second": 5.076,
+      "step": 8910
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.380879282951355,
+      "learning_rate": 5.074626865671642e-06,
+      "loss": 2.0402,
+      "step": 9405
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 2.7763009071350098,
+      "eval_runtime": 24.6683,
+      "eval_samples_per_second": 1289.957,
+      "eval_steps_per_second": 5.067,
+      "step": 9405
+    }
+  ],
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.654827761664e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-9900/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02d893634c4cf925a526dc0ee639631fb8d87a34e1ccc97de7c4baa28df56b6a
+size 5969