Training in progress, step 7500, checkpoint

checkpoint-7500/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-7500/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-7500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74150a13c4781721d9c70858ec71673cbc6282e9819740e7e3685261c7576f93
+size 435544704

checkpoint-7500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81fc8388b723378e23b44c3abcf9f8f7c0a0e68734ddcaa9a7862df5509d1c79
+size 871183627

checkpoint-7500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00f94b51b46c77de2ae03635fa3f0bc465e34461ff844560c09c4cc171e9f3a4
+size 14645

checkpoint-7500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:642827d515e7c3d88d20c04224822701abbbc9250deb1d366045f974d08b8258
+size 1383

checkpoint-7500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca9b8a5501507d40ffae7081184eb5717a9cb61e1a824d7d0f7f2087efd5ab69
+size 1465

checkpoint-7500/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-7500/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-7500/trainer_state.json

@@ -0,0 +1,311 @@
+{
+  "best_global_step": 5200,
+  "best_metric": 5.3865814208984375,
+  "best_model_checkpoint": null,
+  "epoch": 18.75,
+  "eval_steps": 500,
+  "global_step": 7500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025,
+      "grad_norm": 15.434585571289062,
+      "learning_rate": 0.0,
+      "loss": 10.3861,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.4324626922607422,
+      "learning_rate": 7.98e-05,
+      "loss": 7.0655,
+      "step": 400
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.195863723754883,
+      "eval_runtime": 15.8913,
+      "eval_samples_per_second": 1616.543,
+      "eval_steps_per_second": 6.356,
+      "step": 400
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.2896528244018555,
+      "learning_rate": 9.600267379679145e-05,
+      "loss": 5.961,
+      "step": 800
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.855167865753174,
+      "eval_runtime": 15.8141,
+      "eval_samples_per_second": 1624.433,
+      "eval_steps_per_second": 6.387,
+      "step": 800
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.3000872135162354,
+      "learning_rate": 9.065508021390375e-05,
+      "loss": 5.7374,
+      "step": 1200
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.720485210418701,
+      "eval_runtime": 15.9219,
+      "eval_samples_per_second": 1613.439,
+      "eval_steps_per_second": 6.343,
+      "step": 1200
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.3779990673065186,
+      "learning_rate": 8.530748663101604e-05,
+      "loss": 5.6049,
+      "step": 1600
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 5.622853755950928,
+      "eval_runtime": 15.9133,
+      "eval_samples_per_second": 1614.309,
+      "eval_steps_per_second": 6.347,
+      "step": 1600
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.3654310703277588,
+      "learning_rate": 7.995989304812834e-05,
+      "loss": 5.498,
+      "step": 2000
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 5.553117752075195,
+      "eval_runtime": 15.8859,
+      "eval_samples_per_second": 1617.098,
+      "eval_steps_per_second": 6.358,
+      "step": 2000
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.5499093532562256,
+      "learning_rate": 7.461229946524065e-05,
+      "loss": 5.4056,
+      "step": 2400
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.501069068908691,
+      "eval_runtime": 16.1578,
+      "eval_samples_per_second": 1589.881,
+      "eval_steps_per_second": 6.251,
+      "step": 2400
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.5300403833389282,
+      "learning_rate": 6.926470588235294e-05,
+      "loss": 5.3232,
+      "step": 2800
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.458910942077637,
+      "eval_runtime": 15.935,
+      "eval_samples_per_second": 1612.116,
+      "eval_steps_per_second": 6.338,
+      "step": 2800
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.7089910507202148,
+      "learning_rate": 6.391711229946524e-05,
+      "loss": 5.2493,
+      "step": 3200
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.4299397468566895,
+      "eval_runtime": 15.9035,
+      "eval_samples_per_second": 1615.306,
+      "eval_steps_per_second": 6.351,
+      "step": 3200
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.7956774234771729,
+      "learning_rate": 5.8569518716577545e-05,
+      "loss": 5.1807,
+      "step": 3600
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.408169746398926,
+      "eval_runtime": 15.9504,
+      "eval_samples_per_second": 1610.556,
+      "eval_steps_per_second": 6.332,
+      "step": 3600
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.8985629081726074,
+      "learning_rate": 5.322192513368984e-05,
+      "loss": 5.1166,
+      "step": 4000
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.396389484405518,
+      "eval_runtime": 15.8668,
+      "eval_samples_per_second": 1619.042,
+      "eval_steps_per_second": 6.365,
+      "step": 4000
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 2.1109752655029297,
+      "learning_rate": 4.787433155080214e-05,
+      "loss": 5.0565,
+      "step": 4400
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.391461372375488,
+      "eval_runtime": 15.9444,
+      "eval_samples_per_second": 1611.158,
+      "eval_steps_per_second": 6.334,
+      "step": 4400
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 2.173975706100464,
+      "learning_rate": 4.2526737967914444e-05,
+      "loss": 4.9988,
+      "step": 4800
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.386951446533203,
+      "eval_runtime": 15.9572,
+      "eval_samples_per_second": 1609.873,
+      "eval_steps_per_second": 6.329,
+      "step": 4800
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 2.42020320892334,
+      "learning_rate": 3.717914438502674e-05,
+      "loss": 4.9451,
+      "step": 5200
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.3865814208984375,
+      "eval_runtime": 15.9144,
+      "eval_samples_per_second": 1614.195,
+      "eval_steps_per_second": 6.346,
+      "step": 5200
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 2.445915460586548,
+      "learning_rate": 3.1831550802139034e-05,
+      "loss": 4.8937,
+      "step": 5600
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.3876051902771,
+      "eval_runtime": 15.9181,
+      "eval_samples_per_second": 1613.821,
+      "eval_steps_per_second": 6.345,
+      "step": 5600
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 2.670419931411743,
+      "learning_rate": 2.6483957219251336e-05,
+      "loss": 4.8481,
+      "step": 6000
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.392692565917969,
+      "eval_runtime": 15.8838,
+      "eval_samples_per_second": 1617.305,
+      "eval_steps_per_second": 6.359,
+      "step": 6000
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 2.802830696105957,
+      "learning_rate": 2.1136363636363638e-05,
+      "loss": 4.8086,
+      "step": 6400
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.402634143829346,
+      "eval_runtime": 15.9917,
+      "eval_samples_per_second": 1606.394,
+      "eval_steps_per_second": 6.316,
+      "step": 6400
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 2.862973213195801,
+      "learning_rate": 1.5788770053475936e-05,
+      "loss": 4.7739,
+      "step": 6800
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 5.406018257141113,
+      "eval_runtime": 16.0183,
+      "eval_samples_per_second": 1603.731,
+      "eval_steps_per_second": 6.305,
+      "step": 6800
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 3.028500556945801,
+      "learning_rate": 1.0441176470588236e-05,
+      "loss": 4.7446,
+      "step": 7200
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 5.41143274307251,
+      "eval_runtime": 15.9098,
+      "eval_samples_per_second": 1614.669,
+      "eval_steps_per_second": 6.348,
+      "step": 7200
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8000,
+  "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": 1.25169726947328e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dac463600d9699a5f5710b48ecc0e815d9a4d7122a6e0126282ae7df64e105a
+size 5905