Training in progress, step 9000, checkpoint

checkpoint-9000/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-9000/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-9000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:023368dc44139f595d803ea2bc6592b728e5e734d48179e341505c56bd516435
+size 435544704

checkpoint-9000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:077d77dda765419b247991146ac100b48d65ebbd4abb1deb4fc184cefab90373
+size 871183627

checkpoint-9000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c43a92a56a9d3b78795be183b356bf57036cd02db33265682df95f1fbcd70f4f
+size 14709

checkpoint-9000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00b4eb0c27b559cfd649f7c80a6d3394a55880d92b3ee2ca98f71b67e7b2333a
+size 1383

checkpoint-9000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:107f2d80a79386f72a27449209cbd620bfb865edc2cac160f728ad4150c2c03b
+size 1465

checkpoint-9000/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-9000/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-9000/trainer_state.json

@@ -0,0 +1,311 @@
+{
+  "best_global_step": 4960,
+  "best_metric": 2.707667589187622,
+  "best_model_checkpoint": null,
+  "epoch": 18.14516129032258,
+  "eval_steps": 500,
+  "global_step": 9000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0020161290322580645,
+      "grad_norm": 29.451557159423828,
+      "learning_rate": 0.0,
+      "loss": 10.3668,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.9898180961608887,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 4.4948,
+      "step": 496
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.2884881496429443,
+      "eval_runtime": 19.6318,
+      "eval_samples_per_second": 1622.471,
+      "eval_steps_per_second": 6.367,
+      "step": 496
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.733898401260376,
+      "learning_rate": 9.477659574468086e-05,
+      "loss": 3.0395,
+      "step": 992
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.9790470600128174,
+      "eval_runtime": 19.6746,
+      "eval_samples_per_second": 1618.936,
+      "eval_steps_per_second": 6.353,
+      "step": 992
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7452684640884399,
+      "learning_rate": 8.950000000000001e-05,
+      "loss": 2.8091,
+      "step": 1488
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.8684024810791016,
+      "eval_runtime": 19.5669,
+      "eval_samples_per_second": 1627.847,
+      "eval_steps_per_second": 6.388,
+      "step": 1488
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7501894235610962,
+      "learning_rate": 8.422340425531914e-05,
+      "loss": 2.6886,
+      "step": 1984
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.807705879211426,
+      "eval_runtime": 19.9207,
+      "eval_samples_per_second": 1598.937,
+      "eval_steps_per_second": 6.275,
+      "step": 1984
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7561497688293457,
+      "learning_rate": 7.894680851063831e-05,
+      "loss": 2.6051,
+      "step": 2480
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.769815444946289,
+      "eval_runtime": 19.6882,
+      "eval_samples_per_second": 1617.821,
+      "eval_steps_per_second": 6.349,
+      "step": 2480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.7271246910095215,
+      "learning_rate": 7.367021276595744e-05,
+      "loss": 2.5391,
+      "step": 2976
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.7428948879241943,
+      "eval_runtime": 19.716,
+      "eval_samples_per_second": 1615.54,
+      "eval_steps_per_second": 6.34,
+      "step": 2976
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.7691565752029419,
+      "learning_rate": 6.83936170212766e-05,
+      "loss": 2.4829,
+      "step": 3472
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.7244672775268555,
+      "eval_runtime": 19.6939,
+      "eval_samples_per_second": 1617.353,
+      "eval_steps_per_second": 6.347,
+      "step": 3472
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.8312140703201294,
+      "learning_rate": 6.311702127659575e-05,
+      "loss": 2.4319,
+      "step": 3968
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.7130935192108154,
+      "eval_runtime": 19.9468,
+      "eval_samples_per_second": 1596.851,
+      "eval_steps_per_second": 6.267,
+      "step": 3968
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.8690492510795593,
+      "learning_rate": 5.7840425531914896e-05,
+      "loss": 2.3847,
+      "step": 4464
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.709930896759033,
+      "eval_runtime": 19.7333,
+      "eval_samples_per_second": 1614.127,
+      "eval_steps_per_second": 6.334,
+      "step": 4464
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.894021213054657,
+      "learning_rate": 5.256382978723404e-05,
+      "loss": 2.3401,
+      "step": 4960
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.707667589187622,
+      "eval_runtime": 19.7146,
+      "eval_samples_per_second": 1615.655,
+      "eval_steps_per_second": 6.34,
+      "step": 4960
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.9958030581474304,
+      "learning_rate": 4.728723404255319e-05,
+      "loss": 2.2975,
+      "step": 5456
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.710210084915161,
+      "eval_runtime": 19.6907,
+      "eval_samples_per_second": 1617.615,
+      "eval_steps_per_second": 6.348,
+      "step": 5456
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.0411185026168823,
+      "learning_rate": 4.201063829787234e-05,
+      "loss": 2.2563,
+      "step": 5952
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.715811252593994,
+      "eval_runtime": 19.7014,
+      "eval_samples_per_second": 1616.739,
+      "eval_steps_per_second": 6.345,
+      "step": 5952
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.0888656377792358,
+      "learning_rate": 3.673404255319149e-05,
+      "loss": 2.2172,
+      "step": 6448
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 2.7241718769073486,
+      "eval_runtime": 19.7157,
+      "eval_samples_per_second": 1615.565,
+      "eval_steps_per_second": 6.34,
+      "step": 6448
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.185387134552002,
+      "learning_rate": 3.1457446808510635e-05,
+      "loss": 2.18,
+      "step": 6944
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 2.734015703201294,
+      "eval_runtime": 19.7221,
+      "eval_samples_per_second": 1615.043,
+      "eval_steps_per_second": 6.338,
+      "step": 6944
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.262649655342102,
+      "learning_rate": 2.6180851063829785e-05,
+      "loss": 2.1439,
+      "step": 7440
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.7426507472991943,
+      "eval_runtime": 19.7,
+      "eval_samples_per_second": 1616.85,
+      "eval_steps_per_second": 6.345,
+      "step": 7440
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.3151895999908447,
+      "learning_rate": 2.090425531914894e-05,
+      "loss": 2.1121,
+      "step": 7936
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 2.753204584121704,
+      "eval_runtime": 19.7208,
+      "eval_samples_per_second": 1615.144,
+      "eval_steps_per_second": 6.338,
+      "step": 7936
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.3545297384262085,
+      "learning_rate": 1.5627659574468086e-05,
+      "loss": 2.0841,
+      "step": 8432
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 2.762556552886963,
+      "eval_runtime": 19.7442,
+      "eval_samples_per_second": 1613.233,
+      "eval_steps_per_second": 6.331,
+      "step": 8432
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.3998233079910278,
+      "learning_rate": 1.0351063829787235e-05,
+      "loss": 2.06,
+      "step": 8928
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 2.771113157272339,
+      "eval_runtime": 19.6908,
+      "eval_samples_per_second": 1617.612,
+      "eval_steps_per_second": 6.348,
+      "step": 8928
+    }
+  ],
+  "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": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.50419551813632e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-9000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a430d46d37b243b748679f1634186779832282a94872383096add239d05df62d
+size 5905