Training in progress, step 7980, checkpoint

checkpoint-7980/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-7980/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-7980/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5080a487dbe2ef9c324898c5263e8c426ba33488d69927a71edfb7140b602095
+size 435544704

checkpoint-7980/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d65dfe50283d48643f8f4134f7af9f2e56d9a70d70fd8f78e198311c6e785c50
+size 871183627

checkpoint-7980/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85e92aedf3bb2b584b82aa5652c113549c2c88ebc9cf5bfcb8bdf861d6da517f
+size 14645

checkpoint-7980/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd66dab9e9319af067434178d43a0f65cb09e43fb4159f93f4611d07cb83f8b4
+size 1383

checkpoint-7980/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63bf0b640c93a96c51c4d80210bc0946fee83c495c5efd8b2af11c6dab457105
+size 1465

checkpoint-7980/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-7980/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-7980/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 6783,
+  "best_metric": 4.3653178215026855,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 7980,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002506265664160401,
+      "grad_norm": 15.463517189025879,
+      "learning_rate": 0.0,
+      "loss": 10.3917,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.1639904975891113,
+      "learning_rate": 7.960000000000001e-05,
+      "loss": 6.9863,
+      "step": 399
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.981646537780762,
+      "eval_runtime": 20.2663,
+      "eval_samples_per_second": 1263.084,
+      "eval_steps_per_second": 4.934,
+      "step": 399
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.2706925868988037,
+      "learning_rate": 9.601876675603218e-05,
+      "loss": 5.6184,
+      "step": 798
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.361081123352051,
+      "eval_runtime": 20.2304,
+      "eval_samples_per_second": 1265.322,
+      "eval_steps_per_second": 4.943,
+      "step": 798
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.1981648206710815,
+      "learning_rate": 9.067024128686328e-05,
+      "loss": 5.168,
+      "step": 1197
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.063942909240723,
+      "eval_runtime": 20.2184,
+      "eval_samples_per_second": 1266.077,
+      "eval_steps_per_second": 4.946,
+      "step": 1197
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.2216662168502808,
+      "learning_rate": 8.532171581769438e-05,
+      "loss": 4.9038,
+      "step": 1596
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.871644496917725,
+      "eval_runtime": 20.2389,
+      "eval_samples_per_second": 1264.79,
+      "eval_steps_per_second": 4.941,
+      "step": 1596
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2172868251800537,
+      "learning_rate": 7.997319034852548e-05,
+      "loss": 4.7112,
+      "step": 1995
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.736544609069824,
+      "eval_runtime": 20.2297,
+      "eval_samples_per_second": 1265.367,
+      "eval_steps_per_second": 4.943,
+      "step": 1995
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.2852596044540405,
+      "learning_rate": 7.462466487935658e-05,
+      "loss": 4.5567,
+      "step": 2394
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.636835098266602,
+      "eval_runtime": 16.0545,
+      "eval_samples_per_second": 1594.447,
+      "eval_steps_per_second": 6.229,
+      "step": 2394
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.3125325441360474,
+      "learning_rate": 6.927613941018766e-05,
+      "loss": 4.4271,
+      "step": 2793
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.561600685119629,
+      "eval_runtime": 15.8555,
+      "eval_samples_per_second": 1614.451,
+      "eval_steps_per_second": 6.307,
+      "step": 2793
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.3965082168579102,
+      "learning_rate": 6.392761394101878e-05,
+      "loss": 4.3154,
+      "step": 3192
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.50303316116333,
+      "eval_runtime": 15.82,
+      "eval_samples_per_second": 1618.076,
+      "eval_steps_per_second": 6.321,
+      "step": 3192
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.4360967874526978,
+      "learning_rate": 5.8579088471849864e-05,
+      "loss": 4.2166,
+      "step": 3591
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.459570407867432,
+      "eval_runtime": 15.8794,
+      "eval_samples_per_second": 1612.028,
+      "eval_steps_per_second": 6.297,
+      "step": 3591
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.5055606365203857,
+      "learning_rate": 5.3230563002680965e-05,
+      "loss": 4.1296,
+      "step": 3990
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.429525852203369,
+      "eval_runtime": 15.9104,
+      "eval_samples_per_second": 1608.889,
+      "eval_steps_per_second": 6.285,
+      "step": 3990
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.6428686380386353,
+      "learning_rate": 4.7882037533512065e-05,
+      "loss": 4.0523,
+      "step": 4389
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.406736373901367,
+      "eval_runtime": 15.8732,
+      "eval_samples_per_second": 1612.655,
+      "eval_steps_per_second": 6.3,
+      "step": 4389
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.722487211227417,
+      "learning_rate": 4.2533512064343165e-05,
+      "loss": 3.9838,
+      "step": 4788
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.389580726623535,
+      "eval_runtime": 15.877,
+      "eval_samples_per_second": 1612.274,
+      "eval_steps_per_second": 6.298,
+      "step": 4788
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.8150577545166016,
+      "learning_rate": 3.7184986595174266e-05,
+      "loss": 3.9219,
+      "step": 5187
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.378135681152344,
+      "eval_runtime": 15.8421,
+      "eval_samples_per_second": 1615.821,
+      "eval_steps_per_second": 6.312,
+      "step": 5187
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.8721543550491333,
+      "learning_rate": 3.1836461126005366e-05,
+      "loss": 3.8667,
+      "step": 5586
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.3692450523376465,
+      "eval_runtime": 15.8491,
+      "eval_samples_per_second": 1615.103,
+      "eval_steps_per_second": 6.309,
+      "step": 5586
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.9306551218032837,
+      "learning_rate": 2.648793565683646e-05,
+      "loss": 3.8168,
+      "step": 5985
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.36772346496582,
+      "eval_runtime": 15.8933,
+      "eval_samples_per_second": 1610.616,
+      "eval_steps_per_second": 6.292,
+      "step": 5985
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 2.002411127090454,
+      "learning_rate": 2.113941018766756e-05,
+      "loss": 3.7728,
+      "step": 6384
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.367446422576904,
+      "eval_runtime": 15.8568,
+      "eval_samples_per_second": 1614.322,
+      "eval_steps_per_second": 6.306,
+      "step": 6384
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 2.0900936126708984,
+      "learning_rate": 1.579088471849866e-05,
+      "loss": 3.7344,
+      "step": 6783
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.3653178215026855,
+      "eval_runtime": 15.8763,
+      "eval_samples_per_second": 1612.34,
+      "eval_steps_per_second": 6.299,
+      "step": 6783
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 2.0844414234161377,
+      "learning_rate": 1.044235924932976e-05,
+      "loss": 3.7016,
+      "step": 7182
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 4.36598539352417,
+      "eval_runtime": 15.811,
+      "eval_samples_per_second": 1618.999,
+      "eval_steps_per_second": 6.325,
+      "step": 7182
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 2.1372909545898438,
+      "learning_rate": 5.093833780160858e-06,
+      "loss": 3.6741,
+      "step": 7581
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 4.3677167892456055,
+      "eval_runtime": 15.9043,
+      "eval_samples_per_second": 1609.5,
+      "eval_steps_per_second": 6.288,
+      "step": 7581
+    }
+  ],
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.332654686208e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7980/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea117d68be7d0404164110db493aacb08f9f2f28e006665b40d4696c7c8404fa
+size 5905