Training in progress, step 7820, checkpoint

checkpoint-7820/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-7820/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-7820/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:edbbff3004b72b6418357bee36c8203b5d4da393f99310fb0d72c1386022ba7f
+size 435544704

checkpoint-7820/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9186a36c6d303db448b3e6d359e573023e4334e3f570741d6e109a2f08abe194
+size 871183627

checkpoint-7820/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9340b432fa5a8b2e9ac3cde275c0672d7e7d3ed61f94acf2ac2a1956274058c8
+size 14709

checkpoint-7820/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f85d4bdba50d945e851aeab4d6c35670010d85c51eadcea7ad7fc231827fd2d
+size 1383

checkpoint-7820/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2228ca91d60f4d5fbaa26237c5737b472ba097cd903e9a7edb7eca80a503d063
+size 1465

checkpoint-7820/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-7820/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-7820/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 6256,
+  "best_metric": 4.33682107925415,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 7820,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025575447570332483,
+      "grad_norm": 15.82417106628418,
+      "learning_rate": 0.0,
+      "loss": 10.3384,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.0819566249847412,
+      "learning_rate": 7.800000000000001e-05,
+      "loss": 6.8358,
+      "step": 391
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.744270324707031,
+      "eval_runtime": 19.257,
+      "eval_samples_per_second": 1305.497,
+      "eval_steps_per_second": 5.141,
+      "step": 391
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9555849432945251,
+      "learning_rate": 9.615068493150685e-05,
+      "loss": 5.4018,
+      "step": 782
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.183175563812256,
+      "eval_runtime": 19.2203,
+      "eval_samples_per_second": 1307.992,
+      "eval_steps_per_second": 5.151,
+      "step": 782
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9064885973930359,
+      "learning_rate": 9.07945205479452e-05,
+      "loss": 5.0084,
+      "step": 1173
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.933381080627441,
+      "eval_runtime": 19.2736,
+      "eval_samples_per_second": 1304.376,
+      "eval_steps_per_second": 5.137,
+      "step": 1173
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.9678681492805481,
+      "learning_rate": 8.543835616438357e-05,
+      "loss": 4.784,
+      "step": 1564
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.774353504180908,
+      "eval_runtime": 19.298,
+      "eval_samples_per_second": 1302.727,
+      "eval_steps_per_second": 5.13,
+      "step": 1564
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.9735383987426758,
+      "learning_rate": 8.008219178082192e-05,
+      "loss": 4.6185,
+      "step": 1955
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.660486221313477,
+      "eval_runtime": 19.3238,
+      "eval_samples_per_second": 1300.989,
+      "eval_steps_per_second": 5.123,
+      "step": 1955
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.9800556302070618,
+      "learning_rate": 7.472602739726028e-05,
+      "loss": 4.484,
+      "step": 2346
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.573859214782715,
+      "eval_runtime": 19.3847,
+      "eval_samples_per_second": 1296.9,
+      "eval_steps_per_second": 5.107,
+      "step": 2346
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.0170916318893433,
+      "learning_rate": 6.936986301369864e-05,
+      "loss": 4.3687,
+      "step": 2737
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.506965637207031,
+      "eval_runtime": 19.3564,
+      "eval_samples_per_second": 1298.798,
+      "eval_steps_per_second": 5.115,
+      "step": 2737
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.0701326131820679,
+      "learning_rate": 6.401369863013699e-05,
+      "loss": 4.2672,
+      "step": 3128
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.455434799194336,
+      "eval_runtime": 19.352,
+      "eval_samples_per_second": 1299.092,
+      "eval_steps_per_second": 5.116,
+      "step": 3128
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.1530416011810303,
+      "learning_rate": 5.865753424657534e-05,
+      "loss": 4.1753,
+      "step": 3519
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.41667366027832,
+      "eval_runtime": 19.3007,
+      "eval_samples_per_second": 1302.542,
+      "eval_steps_per_second": 5.129,
+      "step": 3519
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.2036124467849731,
+      "learning_rate": 5.330136986301371e-05,
+      "loss": 4.0922,
+      "step": 3910
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.388204574584961,
+      "eval_runtime": 19.3087,
+      "eval_samples_per_second": 1302.006,
+      "eval_steps_per_second": 5.127,
+      "step": 3910
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.2782891988754272,
+      "learning_rate": 4.794520547945205e-05,
+      "loss": 4.0179,
+      "step": 4301
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.36767053604126,
+      "eval_runtime": 19.3884,
+      "eval_samples_per_second": 1296.653,
+      "eval_steps_per_second": 5.106,
+      "step": 4301
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.362265706062317,
+      "learning_rate": 4.258904109589041e-05,
+      "loss": 3.9516,
+      "step": 4692
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.352928161621094,
+      "eval_runtime": 19.5879,
+      "eval_samples_per_second": 1283.446,
+      "eval_steps_per_second": 5.054,
+      "step": 4692
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.39359450340271,
+      "learning_rate": 3.723287671232877e-05,
+      "loss": 3.8915,
+      "step": 5083
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.344639778137207,
+      "eval_runtime": 19.346,
+      "eval_samples_per_second": 1299.492,
+      "eval_steps_per_second": 5.117,
+      "step": 5083
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4617666006088257,
+      "learning_rate": 3.187671232876712e-05,
+      "loss": 3.8371,
+      "step": 5474
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.3404436111450195,
+      "eval_runtime": 19.3581,
+      "eval_samples_per_second": 1298.68,
+      "eval_steps_per_second": 5.114,
+      "step": 5474
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.5229603052139282,
+      "learning_rate": 2.6520547945205482e-05,
+      "loss": 3.7888,
+      "step": 5865
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.337045192718506,
+      "eval_runtime": 19.2969,
+      "eval_samples_per_second": 1302.799,
+      "eval_steps_per_second": 5.13,
+      "step": 5865
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.5618113279342651,
+      "learning_rate": 2.1164383561643834e-05,
+      "loss": 3.7454,
+      "step": 6256
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.33682107925415,
+      "eval_runtime": 19.3114,
+      "eval_samples_per_second": 1301.821,
+      "eval_steps_per_second": 5.127,
+      "step": 6256
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.6074235439300537,
+      "learning_rate": 1.5808219178082193e-05,
+      "loss": 3.7072,
+      "step": 6647
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.338222503662109,
+      "eval_runtime": 19.2556,
+      "eval_samples_per_second": 1305.597,
+      "eval_steps_per_second": 5.141,
+      "step": 6647
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.6486425399780273,
+      "learning_rate": 1.0452054794520549e-05,
+      "loss": 3.6747,
+      "step": 7038
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 4.340147495269775,
+      "eval_runtime": 19.5515,
+      "eval_samples_per_second": 1285.836,
+      "eval_steps_per_second": 5.064,
+      "step": 7038
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.6665706634521484,
+      "learning_rate": 5.095890410958904e-06,
+      "loss": 3.6475,
+      "step": 7429
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 4.341259479522705,
+      "eval_runtime": 19.3292,
+      "eval_samples_per_second": 1300.624,
+      "eval_steps_per_second": 5.122,
+      "step": 7429
+    }
+  ],
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3068782272512e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7820/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29e16aa618eade0e1c5ca5f51a0bd311a6e3b51408b3fac30e15b32c9441a497
+size 5905