ptsv/gpt2-m_simplewiki_upsampled_tom

README.md

@@ -0,0 +1,67 @@
+---
+library_name: transformers
+license: mit
+base_model: openai-community/gpt2-medium
+tags:
+- generated_from_trainer
+model-index:
+- name: simplewiki_upsampled_tom
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/ptsvil/tom-training/runs/keqgv1un)
+# simplewiki_upsampled_tom
+
+This model is a fine-tuned version of [openai-community/gpt2-medium](https://huggingface.co/openai-community/gpt2-medium) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.1560
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 32
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 256
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 1
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 4.0268        | 0.5387 | 20   | 3.4559          |
+| 3.3888        | 1.0774 | 40   | 3.2204          |
+| 3.2608        | 1.6162 | 60   | 3.1798          |
+| 3.0763        | 2.1549 | 80   | 3.1624          |
+| 3.0912        | 2.6936 | 100  | 3.1531          |
+
+
+### Framework versions
+
+- Transformers 4.44.1
+- Pytorch 2.2.2
+- Datasets 2.18.0
+- Tokenizers 0.19.1

config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "openai-community/gpt2-medium",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 1024,
+  "n_head": 16,
+  "n_inner": null,
+  "n_layer": 24,
+  "n_positions": 1024,
+  "n_special": 0,
+  "predict_special_tokens": true,
+  "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,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.1",
+  "use_cache": true,
+  "vocab_size": 50257
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.44.1"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57dab1e88dcdd06ee121c2247ba36ddd70cb16626d42ace36065d0dce77d3070
+size 1419322880

trainer_state.json

@@ -0,0 +1,483 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.98989898989899,
+  "eval_steps": 20,
+  "global_step": 111,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05387205387205387,
+      "grad_norm": 30.91269874572754,
+      "learning_rate": 0.0001,
+      "loss": 9.4247,
+      "step": 2
+    },
+    {
+      "epoch": 0.10774410774410774,
+      "grad_norm": 17.899776458740234,
+      "learning_rate": 9.818181818181818e-05,
+      "loss": 8.2334,
+      "step": 4
+    },
+    {
+      "epoch": 0.16161616161616163,
+      "grad_norm": 11.289294242858887,
+      "learning_rate": 9.636363636363637e-05,
+      "loss": 6.7049,
+      "step": 6
+    },
+    {
+      "epoch": 0.21548821548821548,
+      "grad_norm": 13.52355670928955,
+      "learning_rate": 9.454545454545455e-05,
+      "loss": 6.0078,
+      "step": 8
+    },
+    {
+      "epoch": 0.26936026936026936,
+      "grad_norm": 6.83944845199585,
+      "learning_rate": 9.272727272727273e-05,
+      "loss": 5.4206,
+      "step": 10
+    },
+    {
+      "epoch": 0.32323232323232326,
+      "grad_norm": 5.749248504638672,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 5.0018,
+      "step": 12
+    },
+    {
+      "epoch": 0.3771043771043771,
+      "grad_norm": 6.119119644165039,
+      "learning_rate": 8.90909090909091e-05,
+      "loss": 4.7585,
+      "step": 14
+    },
+    {
+      "epoch": 0.43097643097643096,
+      "grad_norm": 5.070394992828369,
+      "learning_rate": 8.727272727272727e-05,
+      "loss": 4.4387,
+      "step": 16
+    },
+    {
+      "epoch": 0.48484848484848486,
+      "grad_norm": 5.378049850463867,
+      "learning_rate": 8.545454545454545e-05,
+      "loss": 4.1869,
+      "step": 18
+    },
+    {
+      "epoch": 0.5387205387205387,
+      "grad_norm": 6.204185962677002,
+      "learning_rate": 8.363636363636364e-05,
+      "loss": 4.0268,
+      "step": 20
+    },
+    {
+      "epoch": 0.5387205387205387,
+      "eval_loss": 3.4558703899383545,
+      "eval_runtime": 8.1602,
+      "eval_samples_per_second": 145.708,
+      "eval_steps_per_second": 18.259,
+      "step": 20
+    },
+    {
+      "epoch": 0.5925925925925926,
+      "grad_norm": 4.224447727203369,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": 3.8953,
+      "step": 22
+    },
+    {
+      "epoch": 0.6464646464646465,
+      "grad_norm": 3.6069717407226562,
+      "learning_rate": 8e-05,
+      "loss": 3.8231,
+      "step": 24
+    },
+    {
+      "epoch": 0.7003367003367004,
+      "grad_norm": 2.707709550857544,
+      "learning_rate": 7.818181818181818e-05,
+      "loss": 3.7116,
+      "step": 26
+    },
+    {
+      "epoch": 0.7542087542087542,
+      "grad_norm": 3.31360125541687,
+      "learning_rate": 7.636363636363637e-05,
+      "loss": 3.5999,
+      "step": 28
+    },
+    {
+      "epoch": 0.8080808080808081,
+      "grad_norm": 2.9818968772888184,
+      "learning_rate": 7.454545454545455e-05,
+      "loss": 3.5958,
+      "step": 30
+    },
+    {
+      "epoch": 0.8619528619528619,
+      "grad_norm": 2.68215274810791,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": 3.5518,
+      "step": 32
+    },
+    {
+      "epoch": 0.9158249158249159,
+      "grad_norm": 1.9956828355789185,
+      "learning_rate": 7.090909090909092e-05,
+      "loss": 3.4949,
+      "step": 34
+    },
+    {
+      "epoch": 0.9696969696969697,
+      "grad_norm": 1.9735620021820068,
+      "learning_rate": 6.90909090909091e-05,
+      "loss": 3.4185,
+      "step": 36
+    },
+    {
+      "epoch": 1.0235690235690236,
+      "grad_norm": 1.8688853979110718,
+      "learning_rate": 6.727272727272727e-05,
+      "loss": 3.4284,
+      "step": 38
+    },
+    {
+      "epoch": 1.0774410774410774,
+      "grad_norm": 1.9668887853622437,
+      "learning_rate": 6.545454545454546e-05,
+      "loss": 3.3888,
+      "step": 40
+    },
+    {
+      "epoch": 1.0774410774410774,
+      "eval_loss": 3.2203786373138428,
+      "eval_runtime": 8.1727,
+      "eval_samples_per_second": 145.484,
+      "eval_steps_per_second": 18.231,
+      "step": 40
+    },
+    {
+      "epoch": 1.1313131313131313,
+      "grad_norm": 1.8594480752944946,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": 3.3016,
+      "step": 42
+    },
+    {
+      "epoch": 1.1851851851851851,
+      "grad_norm": 2.0851259231567383,
+      "learning_rate": 6.181818181818182e-05,
+      "loss": 3.2937,
+      "step": 44
+    },
+    {
+      "epoch": 1.239057239057239,
+      "grad_norm": 2.022775411605835,
+      "learning_rate": 6e-05,
+      "loss": 3.3097,
+      "step": 46
+    },
+    {
+      "epoch": 1.2929292929292928,
+      "grad_norm": 1.7249592542648315,
+      "learning_rate": 5.818181818181818e-05,
+      "loss": 3.2856,
+      "step": 48
+    },
+    {
+      "epoch": 1.3468013468013469,
+      "grad_norm": 1.894808292388916,
+      "learning_rate": 5.636363636363636e-05,
+      "loss": 3.2579,
+      "step": 50
+    },
+    {
+      "epoch": 1.4006734006734007,
+      "grad_norm": 2.0816519260406494,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 3.2534,
+      "step": 52
+    },
+    {
+      "epoch": 1.4545454545454546,
+      "grad_norm": 1.8632744550704956,
+      "learning_rate": 5.272727272727272e-05,
+      "loss": 3.2358,
+      "step": 54
+    },
+    {
+      "epoch": 1.5084175084175084,
+      "grad_norm": 1.8851394653320312,
+      "learning_rate": 5.090909090909091e-05,
+      "loss": 3.2256,
+      "step": 56
+    },
+    {
+      "epoch": 1.5622895622895623,
+      "grad_norm": 1.7529875040054321,
+      "learning_rate": 4.909090909090909e-05,
+      "loss": 3.2148,
+      "step": 58
+    },
+    {
+      "epoch": 1.6161616161616161,
+      "grad_norm": 1.8536018133163452,
+      "learning_rate": 4.7272727272727275e-05,
+      "loss": 3.2608,
+      "step": 60
+    },
+    {
+      "epoch": 1.6161616161616161,
+      "eval_loss": 3.1798043251037598,
+      "eval_runtime": 8.1729,
+      "eval_samples_per_second": 145.48,
+      "eval_steps_per_second": 18.231,
+      "step": 60
+    },
+    {
+      "epoch": 1.67003367003367,
+      "grad_norm": 1.9621059894561768,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 3.2202,
+      "step": 62
+    },
+    {
+      "epoch": 1.723905723905724,
+      "grad_norm": 1.9148567914962769,
+      "learning_rate": 4.3636363636363636e-05,
+      "loss": 3.2282,
+      "step": 64
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 1.7804056406021118,
+      "learning_rate": 4.181818181818182e-05,
+      "loss": 3.2654,
+      "step": 66
+    },
+    {
+      "epoch": 1.8316498316498318,
+      "grad_norm": 1.8891396522521973,
+      "learning_rate": 4e-05,
+      "loss": 3.1924,
+      "step": 68
+    },
+    {
+      "epoch": 1.8855218855218854,
+      "grad_norm": 1.8429640531539917,
+      "learning_rate": 3.818181818181819e-05,
+      "loss": 3.1756,
+      "step": 70
+    },
+    {
+      "epoch": 1.9393939393939394,
+      "grad_norm": 1.795774221420288,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 3.2097,
+      "step": 72
+    },
+    {
+      "epoch": 1.9932659932659933,
+      "grad_norm": 1.8442139625549316,
+      "learning_rate": 3.454545454545455e-05,
+      "loss": 3.2271,
+      "step": 74
+    },
+    {
+      "epoch": 2.047138047138047,
+      "grad_norm": 1.6865243911743164,
+      "learning_rate": 3.272727272727273e-05,
+      "loss": 3.1261,
+      "step": 76
+    },
+    {
+      "epoch": 2.101010101010101,
+      "grad_norm": 1.6551978588104248,
+      "learning_rate": 3.090909090909091e-05,
+      "loss": 3.101,
+      "step": 78
+    },
+    {
+      "epoch": 2.154882154882155,
+      "grad_norm": 1.7607771158218384,
+      "learning_rate": 2.909090909090909e-05,
+      "loss": 3.0763,
+      "step": 80
+    },
+    {
+      "epoch": 2.154882154882155,
+      "eval_loss": 3.162412405014038,
+      "eval_runtime": 8.1804,
+      "eval_samples_per_second": 145.348,
+      "eval_steps_per_second": 18.214,
+      "step": 80
+    },
+    {
+      "epoch": 2.208754208754209,
+      "grad_norm": 1.720595359802246,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 3.1138,
+      "step": 82
+    },
+    {
+      "epoch": 2.2626262626262625,
+      "grad_norm": 1.7866605520248413,
+      "learning_rate": 2.5454545454545454e-05,
+      "loss": 3.1239,
+      "step": 84
+    },
+    {
+      "epoch": 2.3164983164983166,
+      "grad_norm": 1.7226550579071045,
+      "learning_rate": 2.3636363636363637e-05,
+      "loss": 3.0789,
+      "step": 86
+    },
+    {
+      "epoch": 2.3703703703703702,
+      "grad_norm": 1.8532516956329346,
+      "learning_rate": 2.1818181818181818e-05,
+      "loss": 3.1286,
+      "step": 88
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "grad_norm": 1.7198731899261475,
+      "learning_rate": 2e-05,
+      "loss": 3.1123,
+      "step": 90
+    },
+    {
+      "epoch": 2.478114478114478,
+      "grad_norm": 1.8436646461486816,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 3.0712,
+      "step": 92
+    },
+    {
+      "epoch": 2.531986531986532,
+      "grad_norm": 1.811488151550293,
+      "learning_rate": 1.6363636363636366e-05,
+      "loss": 3.0727,
+      "step": 94
+    },
+    {
+      "epoch": 2.5858585858585856,
+      "grad_norm": 1.7093764543533325,
+      "learning_rate": 1.4545454545454545e-05,
+      "loss": 3.0966,
+      "step": 96
+    },
+    {
+      "epoch": 2.6397306397306397,
+      "grad_norm": 1.8103761672973633,
+      "learning_rate": 1.2727272727272727e-05,
+      "loss": 3.084,
+      "step": 98
+    },
+    {
+      "epoch": 2.6936026936026938,
+      "grad_norm": 1.8916672468185425,
+      "learning_rate": 1.0909090909090909e-05,
+      "loss": 3.0912,
+      "step": 100
+    },
+    {
+      "epoch": 2.6936026936026938,
+      "eval_loss": 3.153109550476074,
+      "eval_runtime": 8.1741,
+      "eval_samples_per_second": 145.459,
+      "eval_steps_per_second": 18.228,
+      "step": 100
+    },
+    {
+      "epoch": 2.7474747474747474,
+      "grad_norm": 1.7815495729446411,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 3.0743,
+      "step": 102
+    },
+    {
+      "epoch": 2.8013468013468015,
+      "grad_norm": 1.8348023891448975,
+      "learning_rate": 7.272727272727272e-06,
+      "loss": 3.092,
+      "step": 104
+    },
+    {
+      "epoch": 2.855218855218855,
+      "grad_norm": 1.807790756225586,
+      "learning_rate": 5.4545454545454545e-06,
+      "loss": 3.0913,
+      "step": 106
+    },
+    {
+      "epoch": 2.909090909090909,
+      "grad_norm": 1.7084001302719116,
+      "learning_rate": 3.636363636363636e-06,
+      "loss": 3.07,
+      "step": 108
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 1.719460129737854,
+      "learning_rate": 1.818181818181818e-06,
+      "loss": 3.0812,
+      "step": 110
+    },
+    {
+      "epoch": 2.98989898989899,
+      "step": 111,
+      "total_flos": 1.389220151427072e+16,
+      "train_loss": 3.7213772073522344,
+      "train_runtime": 747.8744,
+      "train_samples_per_second": 38.144,
+      "train_steps_per_second": 0.148
+    },
+    {
+      "epoch": 2.98989898989899,
+      "eval_loss": 3.1522939205169678,
+      "eval_runtime": 8.1573,
+      "eval_samples_per_second": 145.759,
+      "eval_steps_per_second": 18.266,
+      "step": 111
+    },
+    {
+      "epoch": 2.98989898989899,
+      "eval_loss": 3.1559650897979736,
+      "eval_runtime": 8.1292,
+      "eval_samples_per_second": 146.262,
+      "eval_steps_per_second": 18.329,
+      "step": 111
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 111,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "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": 1.389220151427072e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbbd4a7d804f089a9e11e47c4c29a70ee4155d91b66f369ec99945c1d2b310ae
+size 5240