Training in progress, step 9920, checkpoint

checkpoint-9920/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-9920/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-9920/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:675d3508fa7cb8e870831ac4c251e91d9c593b7acf2029387e0ed430063c7626
+size 435544704

checkpoint-9920/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddaafd245b185d9d78f3cb888b78b2d0c9f12c1dd5d1e0a9adb7fdb3770f8214
+size 871183627

checkpoint-9920/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c3308a4a74e9e45dc5ae12272d7547693476f29a50259ae24c0bf5da2a0cfef
+size 14709

checkpoint-9920/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf44495ba328110e6e2f85b0f245cd5aa719dd9c1c0e35af59fdae7101f2b6e6
+size 1383

checkpoint-9920/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19c217e250c5a61629bc4291c8017d18fb7655a7ce503917d6671c8ea6d8f3c6
+size 1465

checkpoint-9920/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-9920/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-9920/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 4960,
+  "best_metric": 2.704914093017578,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 9920,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0020161290322580645,
+      "grad_norm": 26.651273727416992,
+      "learning_rate": 0.0,
+      "loss": 10.3824,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.098850131034851,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 4.4925,
+      "step": 496
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.2907111644744873,
+      "eval_runtime": 19.696,
+      "eval_samples_per_second": 1617.185,
+      "eval_steps_per_second": 6.346,
+      "step": 496
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.778118908405304,
+      "learning_rate": 9.477659574468086e-05,
+      "loss": 3.0409,
+      "step": 992
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.979689836502075,
+      "eval_runtime": 19.6744,
+      "eval_samples_per_second": 1618.957,
+      "eval_steps_per_second": 6.353,
+      "step": 992
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7490440607070923,
+      "learning_rate": 8.950000000000001e-05,
+      "loss": 2.8112,
+      "step": 1488
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.8685014247894287,
+      "eval_runtime": 19.6807,
+      "eval_samples_per_second": 1618.438,
+      "eval_steps_per_second": 6.351,
+      "step": 1488
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7581645250320435,
+      "learning_rate": 8.422340425531914e-05,
+      "loss": 2.6901,
+      "step": 1984
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.8072190284729004,
+      "eval_runtime": 19.9442,
+      "eval_samples_per_second": 1597.056,
+      "eval_steps_per_second": 6.267,
+      "step": 1984
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7085155844688416,
+      "learning_rate": 7.894680851063831e-05,
+      "loss": 2.6066,
+      "step": 2480
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.76945161819458,
+      "eval_runtime": 19.6801,
+      "eval_samples_per_second": 1618.484,
+      "eval_steps_per_second": 6.352,
+      "step": 2480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.7407582998275757,
+      "learning_rate": 7.367021276595744e-05,
+      "loss": 2.5405,
+      "step": 2976
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.74255633354187,
+      "eval_runtime": 19.7063,
+      "eval_samples_per_second": 1616.333,
+      "eval_steps_per_second": 6.343,
+      "step": 2976
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.7952260971069336,
+      "learning_rate": 6.83936170212766e-05,
+      "loss": 2.4838,
+      "step": 3472
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.724544048309326,
+      "eval_runtime": 19.7078,
+      "eval_samples_per_second": 1616.214,
+      "eval_steps_per_second": 6.343,
+      "step": 3472
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.8141568303108215,
+      "learning_rate": 6.311702127659575e-05,
+      "loss": 2.4327,
+      "step": 3968
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.715358257293701,
+      "eval_runtime": 19.928,
+      "eval_samples_per_second": 1598.355,
+      "eval_steps_per_second": 6.273,
+      "step": 3968
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.8686069846153259,
+      "learning_rate": 5.7840425531914896e-05,
+      "loss": 2.3855,
+      "step": 4464
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.7087273597717285,
+      "eval_runtime": 19.7166,
+      "eval_samples_per_second": 1615.492,
+      "eval_steps_per_second": 6.34,
+      "step": 4464
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.9183729887008667,
+      "learning_rate": 5.256382978723404e-05,
+      "loss": 2.3404,
+      "step": 4960
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.704914093017578,
+      "eval_runtime": 19.7302,
+      "eval_samples_per_second": 1614.375,
+      "eval_steps_per_second": 6.335,
+      "step": 4960
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.9801062345504761,
+      "learning_rate": 4.728723404255319e-05,
+      "loss": 2.2977,
+      "step": 5456
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.7106637954711914,
+      "eval_runtime": 19.733,
+      "eval_samples_per_second": 1614.151,
+      "eval_steps_per_second": 6.335,
+      "step": 5456
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.037522792816162,
+      "learning_rate": 4.201063829787234e-05,
+      "loss": 2.2562,
+      "step": 5952
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.71606183052063,
+      "eval_runtime": 19.714,
+      "eval_samples_per_second": 1615.707,
+      "eval_steps_per_second": 6.341,
+      "step": 5952
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.0967618227005005,
+      "learning_rate": 3.673404255319149e-05,
+      "loss": 2.217,
+      "step": 6448
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 2.7230417728424072,
+      "eval_runtime": 19.7069,
+      "eval_samples_per_second": 1616.29,
+      "eval_steps_per_second": 6.343,
+      "step": 6448
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.1687383651733398,
+      "learning_rate": 3.1457446808510635e-05,
+      "loss": 2.1794,
+      "step": 6944
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 2.730903148651123,
+      "eval_runtime": 19.7161,
+      "eval_samples_per_second": 1615.536,
+      "eval_steps_per_second": 6.34,
+      "step": 6944
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.2304927110671997,
+      "learning_rate": 2.6180851063829785e-05,
+      "loss": 2.1433,
+      "step": 7440
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.7420129776000977,
+      "eval_runtime": 19.7058,
+      "eval_samples_per_second": 1616.379,
+      "eval_steps_per_second": 6.343,
+      "step": 7440
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.3149621486663818,
+      "learning_rate": 2.090425531914894e-05,
+      "loss": 2.1111,
+      "step": 7936
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 2.753695011138916,
+      "eval_runtime": 19.7123,
+      "eval_samples_per_second": 1615.844,
+      "eval_steps_per_second": 6.341,
+      "step": 7936
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.3354177474975586,
+      "learning_rate": 1.5627659574468086e-05,
+      "loss": 2.083,
+      "step": 8432
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 2.7623157501220703,
+      "eval_runtime": 19.7176,
+      "eval_samples_per_second": 1615.406,
+      "eval_steps_per_second": 6.339,
+      "step": 8432
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.391857624053955,
+      "learning_rate": 1.0351063829787235e-05,
+      "loss": 2.0588,
+      "step": 8928
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 2.76987624168396,
+      "eval_runtime": 19.721,
+      "eval_samples_per_second": 1615.131,
+      "eval_steps_per_second": 6.338,
+      "step": 8928
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.4291868209838867,
+      "learning_rate": 5.0744680851063836e-06,
+      "loss": 2.0383,
+      "step": 9424
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 2.777284860610962,
+      "eval_runtime": 19.6962,
+      "eval_samples_per_second": 1617.163,
+      "eval_steps_per_second": 6.346,
+      "step": 9424
+    }
+  ],
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.6579502014464e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-9920/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80a28f6964e57a6b6c304cd870efb8784265c8b1f5db42d483504309ea764a3c
+size 5905