Upload folder using huggingface_hub

README.md

@@ -0,0 +1,68 @@
+---
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: GPT2_natural_function_67
+  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. -->
+
+# GPT2_natural_function_67
+
+This model was trained from scratch on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.7168
+- Accuracy: 0.3685
+
+## 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: 128
+- eval_batch_size: 128
+- seed: 67
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 10000
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 4.97          | 1.0   | 6422  | 4.7453          | 0.2805   |
+| 4.3346        | 2.0   | 12844 | 4.1934          | 0.3193   |
+| 4.0708        | 3.0   | 19266 | 3.9999          | 0.3376   |
+| 3.9406        | 4.0   | 25688 | 3.9018          | 0.3471   |
+| 3.8415        | 5.0   | 32110 | 3.8403          | 0.3540   |
+| 3.7713        | 6.0   | 38532 | 3.7970          | 0.3587   |
+| 3.7109        | 7.0   | 44954 | 3.7650          | 0.3626   |
+| 3.6557        | 8.0   | 51376 | 3.7405          | 0.3651   |
+| 3.6097        | 9.0   | 57798 | 3.7246          | 0.3677   |
+| 3.5655        | 10.0  | 64220 | 3.7168          | 0.3685   |
+
+
+### Framework versions
+
+- Transformers 4.30.2
+- Pytorch 2.9.1+cu128
+- Datasets 4.1.1
+- Tokenizers 0.13.3

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.36849457909442135,
+    "eval_loss": 3.7167649269104004,
+    "eval_runtime": 2.2733,
+    "eval_samples": 1178,
+    "eval_samples_per_second": 518.178,
+    "eval_steps_per_second": 4.399,
+    "perplexity": 41.13111647535073,
+    "train_loss": 4.052415334677184,
+    "train_runtime": 32698.3408,
+    "train_samples": 821994,
+    "train_samples_per_second": 251.387,
+    "train_steps_per_second": 1.964
+}

config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "models/GPT2_natural_function_67",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 2,
+  "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": 128,
+  "pad_token_id": 1,
+  "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": "4.30.2",
+  "use_cache": true,
+  "vocab_size": 32768
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.36849457909442135,
+    "eval_loss": 3.7167649269104004,
+    "eval_runtime": 2.2733,
+    "eval_samples": 1178,
+    "eval_samples_per_second": 518.178,
+    "eval_steps_per_second": 4.399,
+    "perplexity": 41.13111647535073
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 2,
+  "pad_token_id": 1,
+  "transformers_version": "4.30.2"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a386e6a4234a0319b39684bd13f0b1d48996ee03c7eb334e98ab63f0bfb5fa6e
+size 441329395

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,11 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "token": null,
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 4.052415334677184,
+    "train_runtime": 32698.3408,
+    "train_samples": 821994,
+    "train_samples_per_second": 251.387,
+    "train_steps_per_second": 1.964
+}

trainer_state.json

@@ -0,0 +1,499 @@
+{
+  "best_metric": 3.7167649269104004,
+  "best_model_checkpoint": "models/GPT2_natural_function_67/checkpoint-64220",
+  "epoch": 10.0,
+  "global_step": 64220,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 7.6403,
+      "step": 1000
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 2e-05,
+      "loss": 6.1944,
+      "step": 2000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 3e-05,
+      "loss": 5.7388,
+      "step": 3000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4e-05,
+      "loss": 5.4181,
+      "step": 4000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 5e-05,
+      "loss": 5.1669,
+      "step": 5000
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 6e-05,
+      "loss": 4.97,
+      "step": 6000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.28050345574375357,
+      "eval_loss": 4.745330333709717,
+      "eval_runtime": 2.2437,
+      "eval_samples_per_second": 525.032,
+      "eval_steps_per_second": 4.457,
+      "step": 6422
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 7e-05,
+      "loss": 4.7993,
+      "step": 7000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 8e-05,
+      "loss": 4.6673,
+      "step": 8000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9e-05,
+      "loss": 4.5629,
+      "step": 9000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0001,
+      "loss": 4.4737,
+      "step": 10000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.815750645518259e-05,
+      "loss": 4.399,
+      "step": 11000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.631316857248248e-05,
+      "loss": 4.3346,
+      "step": 12000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.31933879657233,
+      "eval_loss": 4.193398952484131,
+      "eval_runtime": 2.2441,
+      "eval_samples_per_second": 524.934,
+      "eval_steps_per_second": 4.456,
+      "step": 12844
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 9.447067502766507e-05,
+      "loss": 4.2713,
+      "step": 13000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 9.262633714496496e-05,
+      "loss": 4.1959,
+      "step": 14000
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 9.078384360014756e-05,
+      "loss": 4.1675,
+      "step": 15000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 8.893950571744743e-05,
+      "loss": 4.1374,
+      "step": 16000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 8.709701217263003e-05,
+      "loss": 4.1132,
+      "step": 17000
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 8.525267428992992e-05,
+      "loss": 4.0906,
+      "step": 18000
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 8.341018074511251e-05,
+      "loss": 4.0708,
+      "step": 19000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.33757335935724503,
+      "eval_loss": 3.999851703643799,
+      "eval_runtime": 2.2451,
+      "eval_samples_per_second": 524.706,
+      "eval_steps_per_second": 4.454,
+      "step": 19266
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 8.15658428624124e-05,
+      "loss": 4.0029,
+      "step": 20000
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 7.972334931759499e-05,
+      "loss": 3.9806,
+      "step": 21000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 7.787901143489488e-05,
+      "loss": 3.971,
+      "step": 22000
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 7.603651789007746e-05,
+      "loss": 3.961,
+      "step": 23000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 7.419218000737735e-05,
+      "loss": 3.9509,
+      "step": 24000
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 7.234968646255994e-05,
+      "loss": 3.9406,
+      "step": 25000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.34709837840728314,
+      "eval_loss": 3.901829719543457,
+      "eval_runtime": 2.2445,
+      "eval_samples_per_second": 524.829,
+      "eval_steps_per_second": 4.455,
+      "step": 25688
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 7.050534857985983e-05,
+      "loss": 3.9065,
+      "step": 26000
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 6.866469937292512e-05,
+      "loss": 3.8622,
+      "step": 27000
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 6.682036149022502e-05,
+      "loss": 3.8579,
+      "step": 28000
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 6.497602360752491e-05,
+      "loss": 3.8586,
+      "step": 29000
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 6.31316857248248e-05,
+      "loss": 3.8531,
+      "step": 30000
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 6.128919218000738e-05,
+      "loss": 3.8469,
+      "step": 31000
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 5.9444854297307275e-05,
+      "loss": 3.8415,
+      "step": 32000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.35395639212331054,
+      "eval_loss": 3.840256690979004,
+      "eval_runtime": 2.247,
+      "eval_samples_per_second": 524.252,
+      "eval_steps_per_second": 4.45,
+      "step": 32110
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 5.760236075248986e-05,
+      "loss": 3.7757,
+      "step": 33000
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 5.5759867207672446e-05,
+      "loss": 3.776,
+      "step": 34000
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 5.391552932497234e-05,
+      "loss": 3.7774,
+      "step": 35000
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 5.207119144227223e-05,
+      "loss": 3.7758,
+      "step": 36000
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 5.022685355957212e-05,
+      "loss": 3.7736,
+      "step": 37000
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 4.8384360014754705e-05,
+      "loss": 3.7713,
+      "step": 38000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.35874229643196126,
+      "eval_loss": 3.797034978866577,
+      "eval_runtime": 2.2428,
+      "eval_samples_per_second": 525.234,
+      "eval_steps_per_second": 4.459,
+      "step": 38532
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 4.6540022132054597e-05,
+      "loss": 3.734,
+      "step": 39000
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 4.469752858723719e-05,
+      "loss": 3.7046,
+      "step": 40000
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 4.285319070453707e-05,
+      "loss": 3.7108,
+      "step": 41000
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 4.1010697159719665e-05,
+      "loss": 3.7108,
+      "step": 42000
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 3.916635927701956e-05,
+      "loss": 3.7112,
+      "step": 43000
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 3.732202139431944e-05,
+      "loss": 3.7109,
+      "step": 44000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3625523040519765,
+      "eval_loss": 3.76499080657959,
+      "eval_runtime": 2.2423,
+      "eval_samples_per_second": 525.348,
+      "eval_steps_per_second": 4.46,
+      "step": 44954
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 3.547768351161933e-05,
+      "loss": 3.7065,
+      "step": 45000
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 3.3635189966801925e-05,
+      "loss": 3.646,
+      "step": 46000
+    },
+    {
+      "epoch": 7.32,
+      "learning_rate": 3.179085208410181e-05,
+      "loss": 3.6519,
+      "step": 47000
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 2.9948358539284398e-05,
+      "loss": 3.6559,
+      "step": 48000
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 2.810402065658429e-05,
+      "loss": 3.6561,
+      "step": 49000
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 2.6259682773884174e-05,
+      "loss": 3.6566,
+      "step": 50000
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 2.4417189229066766e-05,
+      "loss": 3.6557,
+      "step": 51000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3650722564603024,
+      "eval_loss": 3.740530252456665,
+      "eval_runtime": 2.2409,
+      "eval_samples_per_second": 525.684,
+      "eval_steps_per_second": 4.463,
+      "step": 51376
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.2572851346366654e-05,
+      "loss": 3.6199,
+      "step": 52000
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.0730357801549243e-05,
+      "loss": 3.6032,
+      "step": 53000
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.8886019918849134e-05,
+      "loss": 3.6066,
+      "step": 54000
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.7043526374031723e-05,
+      "loss": 3.6088,
+      "step": 55000
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.5199188491331613e-05,
+      "loss": 3.6074,
+      "step": 56000
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 1.3356694946514203e-05,
+      "loss": 3.6097,
+      "step": 57000
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.36772589334652356,
+      "eval_loss": 3.724553108215332,
+      "eval_runtime": 2.2415,
+      "eval_samples_per_second": 525.553,
+      "eval_steps_per_second": 4.461,
+      "step": 57798
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 1.1512357063814091e-05,
+      "loss": 3.5989,
+      "step": 58000
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.66986351899668e-06,
+      "loss": 3.5648,
+      "step": 59000
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 7.82552563629657e-06,
+      "loss": 3.5665,
+      "step": 60000
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 5.9830320914791595e-06,
+      "loss": 3.568,
+      "step": 61000
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 4.138694208779048e-06,
+      "loss": 3.5691,
+      "step": 62000
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 2.296200663961638e-06,
+      "loss": 3.5672,
+      "step": 63000
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 4.5186278126152714e-07,
+      "loss": 3.5655,
+      "step": 64000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.36849457909442135,
+      "eval_loss": 3.7167649269104004,
+      "eval_runtime": 2.2439,
+      "eval_samples_per_second": 524.989,
+      "eval_steps_per_second": 4.457,
+      "step": 64220
+    },
+    {
+      "epoch": 10.0,
+      "step": 64220,
+      "total_flos": 5.3695120637952e+17,
+      "train_loss": 4.052415334677184,
+      "train_runtime": 32698.3408,
+      "train_samples_per_second": 251.387,
+      "train_steps_per_second": 1.964
+    }
+  ],
+  "max_steps": 64220,
+  "num_train_epochs": 10,
+  "total_flos": 5.3695120637952e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8791830e09e2f808ddcc7ed8c741ca83926f06719e290b57ed9b60c30b339028
+size 4881