Training in progress, step 7000, checkpoint

checkpoint-7000/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-7000/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-7000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c2fcd8af7f1efd72b971298edd4d96d8b759a5cd4f2e82c4e49ded1a912e4f4
+size 435544704

checkpoint-7000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d64dfa3907176a6aa1cef5fd07e4ab2cba32d4b64f90fdf12b29b692947473b2
+size 871183627

checkpoint-7000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e714ffd08532ee9954cc56815add8e3918280dbf292c5ce1bd42585a4a1e1776
+size 14645

checkpoint-7000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be40a03f70bb2a35bb567114fb5ceb05ce1222c8df24b1e1100746c617c5be10
+size 1383

checkpoint-7000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:deb1edafa26b4caaefb6cf165cf402eced9c3b27b58fd3375723ad698e99026a
+size 1465

checkpoint-7000/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-7000/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-7000/trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_global_step": 5865,
+  "best_metric": 4.335602283477783,
+  "best_model_checkpoint": null,
+  "epoch": 17.902813299232736,
+  "eval_steps": 500,
+  "global_step": 7000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025575447570332483,
+      "grad_norm": 15.849249839782715,
+      "learning_rate": 0.0,
+      "loss": 10.3746,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.031598448753357,
+      "learning_rate": 7.800000000000001e-05,
+      "loss": 6.8365,
+      "step": 391
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.74590539932251,
+      "eval_runtime": 19.393,
+      "eval_samples_per_second": 1296.345,
+      "eval_steps_per_second": 5.105,
+      "step": 391
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9619669914245605,
+      "learning_rate": 9.615068493150685e-05,
+      "loss": 5.3979,
+      "step": 782
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.181668758392334,
+      "eval_runtime": 19.3113,
+      "eval_samples_per_second": 1301.828,
+      "eval_steps_per_second": 5.127,
+      "step": 782
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9032778739929199,
+      "learning_rate": 9.07945205479452e-05,
+      "loss": 5.0061,
+      "step": 1173
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.930461883544922,
+      "eval_runtime": 19.4629,
+      "eval_samples_per_second": 1291.685,
+      "eval_steps_per_second": 5.087,
+      "step": 1173
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.9254869222640991,
+      "learning_rate": 8.543835616438357e-05,
+      "loss": 4.7815,
+      "step": 1564
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.774387836456299,
+      "eval_runtime": 19.391,
+      "eval_samples_per_second": 1296.48,
+      "eval_steps_per_second": 5.105,
+      "step": 1564
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.9707768559455872,
+      "learning_rate": 8.008219178082192e-05,
+      "loss": 4.6165,
+      "step": 1955
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.658854007720947,
+      "eval_runtime": 19.4424,
+      "eval_samples_per_second": 1293.051,
+      "eval_steps_per_second": 5.092,
+      "step": 1955
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.9788696765899658,
+      "learning_rate": 7.472602739726028e-05,
+      "loss": 4.4815,
+      "step": 2346
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.57161283493042,
+      "eval_runtime": 19.4724,
+      "eval_samples_per_second": 1291.06,
+      "eval_steps_per_second": 5.084,
+      "step": 2346
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.0454139709472656,
+      "learning_rate": 6.936986301369864e-05,
+      "loss": 4.3662,
+      "step": 2737
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.505728721618652,
+      "eval_runtime": 19.3722,
+      "eval_samples_per_second": 1297.737,
+      "eval_steps_per_second": 5.11,
+      "step": 2737
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.0849652290344238,
+      "learning_rate": 6.401369863013699e-05,
+      "loss": 4.2647,
+      "step": 3128
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.456013202667236,
+      "eval_runtime": 19.4196,
+      "eval_samples_per_second": 1294.567,
+      "eval_steps_per_second": 5.098,
+      "step": 3128
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.142387866973877,
+      "learning_rate": 5.865753424657534e-05,
+      "loss": 4.1728,
+      "step": 3519
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.4147868156433105,
+      "eval_runtime": 19.3974,
+      "eval_samples_per_second": 1296.053,
+      "eval_steps_per_second": 5.104,
+      "step": 3519
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.1628907918930054,
+      "learning_rate": 5.330136986301371e-05,
+      "loss": 4.0896,
+      "step": 3910
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.385285377502441,
+      "eval_runtime": 19.4087,
+      "eval_samples_per_second": 1295.295,
+      "eval_steps_per_second": 5.101,
+      "step": 3910
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.2972906827926636,
+      "learning_rate": 4.794520547945205e-05,
+      "loss": 4.0153,
+      "step": 4301
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.366960525512695,
+      "eval_runtime": 19.4166,
+      "eval_samples_per_second": 1294.767,
+      "eval_steps_per_second": 5.099,
+      "step": 4301
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.3291264772415161,
+      "learning_rate": 4.258904109589041e-05,
+      "loss": 3.9487,
+      "step": 4692
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.35342264175415,
+      "eval_runtime": 19.6756,
+      "eval_samples_per_second": 1277.723,
+      "eval_steps_per_second": 5.032,
+      "step": 4692
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.3737049102783203,
+      "learning_rate": 3.723287671232877e-05,
+      "loss": 3.8886,
+      "step": 5083
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.344508647918701,
+      "eval_runtime": 19.3839,
+      "eval_samples_per_second": 1296.954,
+      "eval_steps_per_second": 5.107,
+      "step": 5083
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4470322132110596,
+      "learning_rate": 3.187671232876712e-05,
+      "loss": 3.8345,
+      "step": 5474
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.340525150299072,
+      "eval_runtime": 19.4296,
+      "eval_samples_per_second": 1293.9,
+      "eval_steps_per_second": 5.095,
+      "step": 5474
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.5098106861114502,
+      "learning_rate": 2.6520547945205482e-05,
+      "loss": 3.786,
+      "step": 5865
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.335602283477783,
+      "eval_runtime": 19.3809,
+      "eval_samples_per_second": 1297.15,
+      "eval_steps_per_second": 5.108,
+      "step": 5865
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.5827183723449707,
+      "learning_rate": 2.1164383561643834e-05,
+      "loss": 3.7426,
+      "step": 6256
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.336056232452393,
+      "eval_runtime": 19.5099,
+      "eval_samples_per_second": 1288.579,
+      "eval_steps_per_second": 5.074,
+      "step": 6256
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.6074565649032593,
+      "learning_rate": 1.5808219178082193e-05,
+      "loss": 3.7044,
+      "step": 6647
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.337679386138916,
+      "eval_runtime": 19.4478,
+      "eval_samples_per_second": 1292.693,
+      "eval_steps_per_second": 5.091,
+      "step": 6647
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7820,
+  "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": 1.16987758903296e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:970bd6d16b2730dba3763fdd1a14419de1f1cb106b82c2fb79c1d1c6012263ba
+size 5905