Training in progress, step 7000, checkpoint

checkpoint-7000/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-7000/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-7000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3d172bb93c074a42e810b006258066d478badc4b151e4d34b89bd77faf26d36
+size 435544704

checkpoint-7000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5515ea5840cbc734603595cb66027c992928d49667c98b79bf5edffe14acc3f6
+size 871183627

checkpoint-7000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3a1219520187b0baa308a5e3fb0ac18e1260f3ab89f453b318f6b21a0df5e4d
+size 14709

checkpoint-7000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be40a03f70bb2a35bb567114fb5ceb05ce1222c8df24b1e1100746c617c5be10
+size 1383

checkpoint-7000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28f64db2c1980ed19749438d6000d120a310f2ece6f00f36cb92c8be840e0ae9
+size 1465

checkpoint-7000/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-7000/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-7000/trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_global_step": 6384,
+  "best_metric": 4.364203929901123,
+  "best_model_checkpoint": null,
+  "epoch": 17.54385964912281,
+  "eval_steps": 500,
+  "global_step": 7000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002506265664160401,
+      "grad_norm": 16.177248001098633,
+      "learning_rate": 0.0,
+      "loss": 10.3857,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.4139729738235474,
+      "learning_rate": 7.960000000000001e-05,
+      "loss": 6.9714,
+      "step": 399
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.984301567077637,
+      "eval_runtime": 19.728,
+      "eval_samples_per_second": 1297.547,
+      "eval_steps_per_second": 5.069,
+      "step": 399
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.1047171354293823,
+      "learning_rate": 9.601876675603218e-05,
+      "loss": 5.6132,
+      "step": 798
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.349092960357666,
+      "eval_runtime": 19.6059,
+      "eval_samples_per_second": 1305.631,
+      "eval_steps_per_second": 5.101,
+      "step": 798
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.1410163640975952,
+      "learning_rate": 9.067024128686328e-05,
+      "loss": 5.159,
+      "step": 1197
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.05531644821167,
+      "eval_runtime": 19.6962,
+      "eval_samples_per_second": 1299.641,
+      "eval_steps_per_second": 5.077,
+      "step": 1197
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.142958641052246,
+      "learning_rate": 8.532171581769438e-05,
+      "loss": 4.8974,
+      "step": 1596
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.867596626281738,
+      "eval_runtime": 19.6239,
+      "eval_samples_per_second": 1304.427,
+      "eval_steps_per_second": 5.096,
+      "step": 1596
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2511125802993774,
+      "learning_rate": 7.997319034852548e-05,
+      "loss": 4.7058,
+      "step": 1995
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.733837127685547,
+      "eval_runtime": 19.6301,
+      "eval_samples_per_second": 1304.018,
+      "eval_steps_per_second": 5.094,
+      "step": 1995
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.2677074670791626,
+      "learning_rate": 7.462466487935658e-05,
+      "loss": 4.5523,
+      "step": 2394
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.6313629150390625,
+      "eval_runtime": 19.8191,
+      "eval_samples_per_second": 1291.584,
+      "eval_steps_per_second": 5.046,
+      "step": 2394
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.3199169635772705,
+      "learning_rate": 6.927613941018766e-05,
+      "loss": 4.4235,
+      "step": 2793
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.556069850921631,
+      "eval_runtime": 15.8869,
+      "eval_samples_per_second": 1611.264,
+      "eval_steps_per_second": 6.294,
+      "step": 2793
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.3809373378753662,
+      "learning_rate": 6.392761394101878e-05,
+      "loss": 4.3123,
+      "step": 3192
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.500065326690674,
+      "eval_runtime": 15.8556,
+      "eval_samples_per_second": 1614.448,
+      "eval_steps_per_second": 6.307,
+      "step": 3192
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.4628700017929077,
+      "learning_rate": 5.8579088471849864e-05,
+      "loss": 4.2136,
+      "step": 3591
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.457473278045654,
+      "eval_runtime": 15.8102,
+      "eval_samples_per_second": 1619.081,
+      "eval_steps_per_second": 6.325,
+      "step": 3591
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.5196454524993896,
+      "learning_rate": 5.3230563002680965e-05,
+      "loss": 4.127,
+      "step": 3990
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.426193714141846,
+      "eval_runtime": 15.8684,
+      "eval_samples_per_second": 1613.138,
+      "eval_steps_per_second": 6.302,
+      "step": 3990
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.6150400638580322,
+      "learning_rate": 4.7882037533512065e-05,
+      "loss": 4.0503,
+      "step": 4389
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.405140399932861,
+      "eval_runtime": 15.8676,
+      "eval_samples_per_second": 1613.229,
+      "eval_steps_per_second": 6.302,
+      "step": 4389
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.6789467334747314,
+      "learning_rate": 4.2533512064343165e-05,
+      "loss": 3.9816,
+      "step": 4788
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.387766361236572,
+      "eval_runtime": 15.8258,
+      "eval_samples_per_second": 1617.486,
+      "eval_steps_per_second": 6.319,
+      "step": 4788
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.753027081489563,
+      "learning_rate": 3.7184986595174266e-05,
+      "loss": 3.92,
+      "step": 5187
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.377025127410889,
+      "eval_runtime": 15.8514,
+      "eval_samples_per_second": 1614.877,
+      "eval_steps_per_second": 6.309,
+      "step": 5187
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.82956063747406,
+      "learning_rate": 3.1836461126005366e-05,
+      "loss": 3.8651,
+      "step": 5586
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.3691086769104,
+      "eval_runtime": 15.7814,
+      "eval_samples_per_second": 1622.034,
+      "eval_steps_per_second": 6.337,
+      "step": 5586
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.9336134195327759,
+      "learning_rate": 2.648793565683646e-05,
+      "loss": 3.8158,
+      "step": 5985
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.3644938468933105,
+      "eval_runtime": 15.8448,
+      "eval_samples_per_second": 1615.542,
+      "eval_steps_per_second": 6.311,
+      "step": 5985
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.9763456583023071,
+      "learning_rate": 2.113941018766756e-05,
+      "loss": 3.7721,
+      "step": 6384
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.364203929901123,
+      "eval_runtime": 15.8844,
+      "eval_samples_per_second": 1611.519,
+      "eval_steps_per_second": 6.295,
+      "step": 6384
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 2.042994976043701,
+      "learning_rate": 1.579088471849866e-05,
+      "loss": 3.7336,
+      "step": 6783
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.36444091796875,
+      "eval_runtime": 20.4863,
+      "eval_samples_per_second": 1249.518,
+      "eval_steps_per_second": 4.881,
+      "step": 6783
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7980,
+  "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": 1.16904472068096e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9246d207a42e79cf06122cec1910493f0a50a9bd8056750a46320eb7d7e02038
+size 5905