Training in progress, step 9900, checkpoint

checkpoint-9900/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-9900/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-9900/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d27815784f144c66b5b32f8a492b00eabd15e301847f284e5e9159f4d63e1c35
+size 435544704

checkpoint-9900/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b9411a7ea90282c02f87836fe64176475a71d9fcef5f9b6b99b1b8e26705322
+size 871183627

checkpoint-9900/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:219ef8083fd0b4a04902b418570912b30efd43b862e0424d641ddaf2555fb3cc
+size 14645

checkpoint-9900/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a5ddab22d8a1c52369c0f9d8d0f6489321ff2d9a19616bd19271f43030cbbc
+size 1383

checkpoint-9900/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8732b31a5837ff2411de273d74dc3972f94fe566013c6ddc3264f1791c73ba4
+size 1465

checkpoint-9900/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-9900/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-9900/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 4950,
+  "best_metric": 3.057368755340576,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 9900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00202020202020202,
+      "grad_norm": 29.595918655395508,
+      "learning_rate": 0.0,
+      "loss": 10.3213,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2207119464874268,
+      "learning_rate": 9.88e-05,
+      "loss": 4.8361,
+      "step": 495
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.7897000312805176,
+      "eval_runtime": 19.6861,
+      "eval_samples_per_second": 1614.843,
+      "eval_steps_per_second": 6.35,
+      "step": 495
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.0401272773742676,
+      "learning_rate": 9.478678038379531e-05,
+      "loss": 3.538,
+      "step": 990
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.4326586723327637,
+      "eval_runtime": 19.7059,
+      "eval_samples_per_second": 1613.223,
+      "eval_steps_per_second": 6.343,
+      "step": 990
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0095438957214355,
+      "learning_rate": 8.950959488272922e-05,
+      "loss": 3.2785,
+      "step": 1485
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.287219285964966,
+      "eval_runtime": 19.5988,
+      "eval_samples_per_second": 1622.034,
+      "eval_steps_per_second": 6.378,
+      "step": 1485
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.0992753505706787,
+      "learning_rate": 8.42324093816631e-05,
+      "loss": 3.1342,
+      "step": 1980
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.207063674926758,
+      "eval_runtime": 19.9631,
+      "eval_samples_per_second": 1592.44,
+      "eval_steps_per_second": 6.262,
+      "step": 1980
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.065517783164978,
+      "learning_rate": 7.895522388059702e-05,
+      "loss": 3.032,
+      "step": 2475
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.151052713394165,
+      "eval_runtime": 19.6853,
+      "eval_samples_per_second": 1614.911,
+      "eval_steps_per_second": 6.35,
+      "step": 2475
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.0753278732299805,
+      "learning_rate": 7.367803837953093e-05,
+      "loss": 2.9514,
+      "step": 2970
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.113241195678711,
+      "eval_runtime": 19.795,
+      "eval_samples_per_second": 1605.965,
+      "eval_steps_per_second": 6.315,
+      "step": 2970
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.1069707870483398,
+      "learning_rate": 6.840085287846483e-05,
+      "loss": 2.8836,
+      "step": 3465
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.090135097503662,
+      "eval_runtime": 19.7634,
+      "eval_samples_per_second": 1608.533,
+      "eval_steps_per_second": 6.325,
+      "step": 3465
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.1921679973602295,
+      "learning_rate": 6.312366737739872e-05,
+      "loss": 2.8231,
+      "step": 3960
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.0735952854156494,
+      "eval_runtime": 19.9524,
+      "eval_samples_per_second": 1593.29,
+      "eval_steps_per_second": 6.265,
+      "step": 3960
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.208296537399292,
+      "learning_rate": 5.784648187633263e-05,
+      "loss": 2.7678,
+      "step": 4455
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.0633397102355957,
+      "eval_runtime": 19.7455,
+      "eval_samples_per_second": 1609.99,
+      "eval_steps_per_second": 6.331,
+      "step": 4455
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.3137807846069336,
+      "learning_rate": 5.256929637526653e-05,
+      "loss": 2.7164,
+      "step": 4950
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.057368755340576,
+      "eval_runtime": 19.795,
+      "eval_samples_per_second": 1605.965,
+      "eval_steps_per_second": 6.315,
+      "step": 4950
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.3745468854904175,
+      "learning_rate": 4.7292110874200426e-05,
+      "loss": 2.6678,
+      "step": 5445
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.0588877201080322,
+      "eval_runtime": 19.7143,
+      "eval_samples_per_second": 1612.535,
+      "eval_steps_per_second": 6.341,
+      "step": 5445
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.4381449222564697,
+      "learning_rate": 4.201492537313433e-05,
+      "loss": 2.6213,
+      "step": 5940
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.0614066123962402,
+      "eval_runtime": 19.6703,
+      "eval_samples_per_second": 1616.141,
+      "eval_steps_per_second": 6.355,
+      "step": 5940
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.5913258790969849,
+      "learning_rate": 3.673773987206823e-05,
+      "loss": 2.5765,
+      "step": 6435
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.0655486583709717,
+      "eval_runtime": 19.7216,
+      "eval_samples_per_second": 1611.937,
+      "eval_steps_per_second": 6.338,
+      "step": 6435
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.6668187379837036,
+      "learning_rate": 3.1460554371002134e-05,
+      "loss": 2.5352,
+      "step": 6930
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.0753347873687744,
+      "eval_runtime": 19.8126,
+      "eval_samples_per_second": 1604.537,
+      "eval_steps_per_second": 6.309,
+      "step": 6930
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.680243730545044,
+      "learning_rate": 2.6183368869936037e-05,
+      "loss": 2.4983,
+      "step": 7425
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.081855058670044,
+      "eval_runtime": 19.7547,
+      "eval_samples_per_second": 1609.24,
+      "eval_steps_per_second": 6.328,
+      "step": 7425
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.7872483730316162,
+      "learning_rate": 2.0906183368869936e-05,
+      "loss": 2.4651,
+      "step": 7920
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.0909264087677,
+      "eval_runtime": 19.8179,
+      "eval_samples_per_second": 1604.109,
+      "eval_steps_per_second": 6.307,
+      "step": 7920
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.8887341022491455,
+      "learning_rate": 1.562899786780384e-05,
+      "loss": 2.4353,
+      "step": 8415
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.102728843688965,
+      "eval_runtime": 19.7265,
+      "eval_samples_per_second": 1611.538,
+      "eval_steps_per_second": 6.337,
+      "step": 8415
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.900124430656433,
+      "learning_rate": 1.035181236673774e-05,
+      "loss": 2.4096,
+      "step": 8910
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 3.1099114418029785,
+      "eval_runtime": 19.7618,
+      "eval_samples_per_second": 1608.66,
+      "eval_steps_per_second": 6.325,
+      "step": 8910
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.9827020168304443,
+      "learning_rate": 5.074626865671642e-06,
+      "loss": 2.3879,
+      "step": 9405
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 3.115513801574707,
+      "eval_runtime": 19.7535,
+      "eval_samples_per_second": 1609.333,
+      "eval_steps_per_second": 6.328,
+      "step": 9405
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9900,
+  "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": 1.6536258183168e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-9900/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1252b3e32d62ce25163ace85e76747f0b7980660b25880bb7fafe8be438ddaad
+size 5905