update model card README.md

README.md

@@ -2,6 +2,8 @@
 license: apache-2.0
 tags:
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: expert-uspto
   results: []
@@ -13,6 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 # expert-uspto
 
 This model is a fine-tuned version of [EleutherAI/pythia-1b-deduped](https://huggingface.co/EleutherAI/pythia-1b-deduped) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.2220
+- Accuracy: 0.5362
 
 ## Model description
 
@@ -33,19 +38,26 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
 - train_batch_size: 1
-- eval_batch_size: 1
+- eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 8
 - gradient_accumulation_steps: 8
 - total_train_batch_size: 64
-- total_eval_batch_size: 8
+- total_eval_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- training_steps: 1
+- training_steps: 1000
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 2.2735        | 0.01  | 200  | 2.2464          | 0.5325   |
+| 2.2557        | 0.01  | 400  | 2.2417          | 0.5331   |
+| 2.2342        | 0.02  | 600  | 2.2342          | 0.5344   |
+| 2.2241        | 0.03  | 800  | 2.2267          | 0.5355   |
+| 2.229         | 0.03  | 1000 | 2.2220          | 0.5362   |
 
 
 ### Framework versions