Training in progress, step 7840, checkpoint

checkpoint-7840/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-7840/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-7840/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b675f8f7103bc21a1105f2339714f4f6ed213140acd90d15a6873b7a577a48d8
+size 435544704

checkpoint-7840/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8f1da61196e9876fa9fdf8fc6d0f0d9c351f38e5a2091a7d21d9e2c70b64741
+size 871183627

checkpoint-7840/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d8b1715e6d36c8bff565bdae49b0047bd6745d25aefe5e31eed04e1a2772cbb
+size 14709

checkpoint-7840/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cec32b315ccd88f013125c5586963f78edf1c2eb49c8520bb8a597ff9a34f034
+size 1383

checkpoint-7840/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b44b2e73cf2c8e552ad9b381f9c97f3ebdd480ce862a34df0cf892839025cf49
+size 1465

checkpoint-7840/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-7840/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-7840/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 5096,
+  "best_metric": 5.531695365905762,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 7840,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002551020408163265,
+      "grad_norm": 16.32339096069336,
+      "learning_rate": 0.0,
+      "loss": 10.3351,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6908162236213684,
+      "learning_rate": 7.82e-05,
+      "loss": 7.1094,
+      "step": 392
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.317445755004883,
+      "eval_runtime": 15.5824,
+      "eval_samples_per_second": 1616.887,
+      "eval_steps_per_second": 6.353,
+      "step": 392
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.7368313074111938,
+      "learning_rate": 9.613387978142077e-05,
+      "loss": 6.0963,
+      "step": 784
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 6.024065017700195,
+      "eval_runtime": 15.5591,
+      "eval_samples_per_second": 1619.314,
+      "eval_steps_per_second": 6.363,
+      "step": 784
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7423638701438904,
+      "learning_rate": 9.077868852459017e-05,
+      "loss": 5.8833,
+      "step": 1176
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.884053707122803,
+      "eval_runtime": 15.6179,
+      "eval_samples_per_second": 1613.212,
+      "eval_steps_per_second": 6.339,
+      "step": 1176
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.707794725894928,
+      "learning_rate": 8.542349726775956e-05,
+      "loss": 5.7525,
+      "step": 1568
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 5.788422584533691,
+      "eval_runtime": 15.5176,
+      "eval_samples_per_second": 1623.64,
+      "eval_steps_per_second": 6.38,
+      "step": 1568
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7633333802223206,
+      "learning_rate": 8.006830601092896e-05,
+      "loss": 5.6482,
+      "step": 1960
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 5.718640327453613,
+      "eval_runtime": 15.5828,
+      "eval_samples_per_second": 1616.851,
+      "eval_steps_per_second": 6.353,
+      "step": 1960
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.7929214239120483,
+      "learning_rate": 7.471311475409837e-05,
+      "loss": 5.5583,
+      "step": 2352
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.6633148193359375,
+      "eval_runtime": 15.6437,
+      "eval_samples_per_second": 1610.551,
+      "eval_steps_per_second": 6.328,
+      "step": 2352
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.8233454823493958,
+      "learning_rate": 6.935792349726776e-05,
+      "loss": 5.4779,
+      "step": 2744
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.620563507080078,
+      "eval_runtime": 15.5999,
+      "eval_samples_per_second": 1615.079,
+      "eval_steps_per_second": 6.346,
+      "step": 2744
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.849190354347229,
+      "learning_rate": 6.400273224043716e-05,
+      "loss": 5.4029,
+      "step": 3136
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.586855888366699,
+      "eval_runtime": 15.6428,
+      "eval_samples_per_second": 1610.644,
+      "eval_steps_per_second": 6.329,
+      "step": 3136
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.9548335671424866,
+      "learning_rate": 5.864754098360656e-05,
+      "loss": 5.3324,
+      "step": 3528
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.562436580657959,
+      "eval_runtime": 15.5065,
+      "eval_samples_per_second": 1624.806,
+      "eval_steps_per_second": 6.384,
+      "step": 3528
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.0412702560424805,
+      "learning_rate": 5.329234972677596e-05,
+      "loss": 5.2657,
+      "step": 3920
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.544801235198975,
+      "eval_runtime": 15.5961,
+      "eval_samples_per_second": 1615.463,
+      "eval_steps_per_second": 6.348,
+      "step": 3920
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.1015349626541138,
+      "learning_rate": 4.793715846994536e-05,
+      "loss": 5.2018,
+      "step": 4312
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.536635398864746,
+      "eval_runtime": 15.6212,
+      "eval_samples_per_second": 1612.869,
+      "eval_steps_per_second": 6.338,
+      "step": 4312
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.1938425302505493,
+      "learning_rate": 4.258196721311475e-05,
+      "loss": 5.1412,
+      "step": 4704
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.531932353973389,
+      "eval_runtime": 15.7939,
+      "eval_samples_per_second": 1595.238,
+      "eval_steps_per_second": 6.268,
+      "step": 4704
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.2612738609313965,
+      "learning_rate": 3.722677595628416e-05,
+      "loss": 5.0841,
+      "step": 5096
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.531695365905762,
+      "eval_runtime": 15.6388,
+      "eval_samples_per_second": 1611.059,
+      "eval_steps_per_second": 6.33,
+      "step": 5096
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.3297182321548462,
+      "learning_rate": 3.1871584699453554e-05,
+      "loss": 5.0297,
+      "step": 5488
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.5333709716796875,
+      "eval_runtime": 15.6204,
+      "eval_samples_per_second": 1612.952,
+      "eval_steps_per_second": 6.338,
+      "step": 5488
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.4006597995758057,
+      "learning_rate": 2.651639344262295e-05,
+      "loss": 4.9798,
+      "step": 5880
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.538441181182861,
+      "eval_runtime": 15.6104,
+      "eval_samples_per_second": 1613.986,
+      "eval_steps_per_second": 6.342,
+      "step": 5880
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.5358235836029053,
+      "learning_rate": 2.116120218579235e-05,
+      "loss": 4.9364,
+      "step": 6272
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.547364234924316,
+      "eval_runtime": 15.5999,
+      "eval_samples_per_second": 1615.073,
+      "eval_steps_per_second": 6.346,
+      "step": 6272
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.5607517957687378,
+      "learning_rate": 1.5806010928961746e-05,
+      "loss": 4.899,
+      "step": 6664
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 5.55307149887085,
+      "eval_runtime": 15.5981,
+      "eval_samples_per_second": 1615.266,
+      "eval_steps_per_second": 6.347,
+      "step": 6664
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.6740516424179077,
+      "learning_rate": 1.0450819672131149e-05,
+      "loss": 4.8676,
+      "step": 7056
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 5.559586524963379,
+      "eval_runtime": 15.6369,
+      "eval_samples_per_second": 1611.249,
+      "eval_steps_per_second": 6.331,
+      "step": 7056
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.6653796434402466,
+      "learning_rate": 5.095628415300547e-06,
+      "loss": 4.8422,
+      "step": 7448
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 5.562194347381592,
+      "eval_runtime": 15.6435,
+      "eval_samples_per_second": 1610.576,
+      "eval_steps_per_second": 6.329,
+      "step": 7448
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7840,
+  "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": 1.3095564705792e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7840/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52cf954a78a3d6fc682ff0c3b6ee4cda69040297e62c644ea0256038484a747a
+size 5905