Edit model card
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

'openai-community/gpt2'

training_args = TrainingArguments( output_dir='ECO_GPT', evaluation_strategy="epoch", learning_rate=2e-5, per_device_train_batch_size=16, per_device_eval_batch_size=16, num_train_epochs=32, weight_decay=0.01, save_strategy="epoch", load_best_model_at_end=True, push_to_hub=True ) Epoch Training Loss Validation Loss Precision Recall F1 Accuracy 1 No log 0.197675 0.285714 0.212743 0.243887 0.959920 2 0.368900 0.159557 0.311457 0.302376 0.306849 0.961646 3 0.368900 0.146986 0.324180 0.344222 0.333901 0.961224 4 0.129100 0.145931 0.340494 0.372300 0.355687 0.960939 5 0.129100 0.144521 0.334234 0.367441 0.350051 0.961286 6 0.098700 0.141342 0.323340 0.393089 0.354819 0.962803 7 0.098700 0.144871 0.319322 0.381479 0.347644 0.962950 8 0.079700 0.147587 0.283885 0.377160 0.323942 0.961504 9 0.079700 0.152369 0.294302 0.355562 0.322044 0.961081 10 0.066600 0.154832 0.304663 0.373920 0.335758 0.962580 11 0.066600 0.160076 0.295521 0.381210 0.332940 0.960837 12 0.056600 0.164543 0.306661 0.370410 0.335534 0.962100 13 0.056600 0.166050 0.331369 0.379590 0.353844 0.963194 14 0.048700 0.169381 0.307859 0.383909 0.341704 0.961958 15 0.048700 0.179932 0.306340 0.380940 0.339591 0.960970 16 0.042500 0.180197 0.313375 0.368790 0.338832 0.962193 17 0.042500 0.186068 0.297326 0.384179 0.335218 0.960623 18 0.037500 0.195882 0.313199 0.377970 0.342550 0.961268 19 0.037500 0.197563 0.304962 0.384989 0.340334 0.960783 20 0.034100 0.201612 0.294698 0.370680 0.328351 0.960765 21 0.034100 0.207106 0.296033 0.382829 0.333883 0.959849 22 0.030900 0.205857 0.293838 0.383639 0.332787 0.960080 23 0.030900 0.211530 0.309654 0.383639 0.342699 0.961041 24 0.028700 0.215475 0.312196 0.381479 0.343378 0.961108 25 0.028700 0.213058 0.311651 0.377700 0.341511 0.961700 26 0.026700 0.216595 0.312333 0.378780 0.342362 0.961508 27 0.026700 0.223365 0.318739 0.379320 0.346400 0.961722 28 0.025400 0.224544 0.312472 0.373380 0.340221 0.961517 29 0.025400 0.225070 0.319539 0.381479 0.347773 0.961878 30 0.024300 0.223071 0.308169 0.375810 0.338645 0.961455 31 0.024300 0.225455 0.306649 0.372300 0.336300 0.961268 32 0.023900 0.224405 0.308620 0.373110 0.337815 0.961384

image/png

Downloads last month
1
Safetensors
Model size
125M params
Tensor type
F32
·
Inference API
This model can be loaded on Inference API (serverless).