Update README.md

README.md

@@ -2,18 +2,18 @@
 library_name: peft
 tags:
 - generated_from_trainer
-base_model: meta-llama/Meta-Llama-3-8B-Instruct
+base_model: meta-llama/Meta-Llama-3-70B
 model-index:
-- name: lora_Meta-Llama-3-70B_llama_wizardllm_sorry_3k_safe_3k_66k_diverse_must_refuse_purely_harmful_declar
+- name: lora_Meta-Llama-3-70B_derta
   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. -->
 
-# lora_Meta-Llama-3-70B_llama_wizardllm_sorry_3k_safe_3k_66k_diverse_must_refuse_purely_harmful_declar
+# lora_Meta-Llama-3-70B_derta
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-70B](https://huggingface.co/meta-llama/Meta-Llama-3-70B) on the Evol-Instruct and BeaverTails dataset.
 
 ## Model description
 
@@ -33,13 +33,13 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 10
+- train_batch_size: 16
 - eval_batch_size: 1
 - seed: 1
 - distributed_type: multi-GPU
-- num_devices: 6
+- num_devices: 8
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 120
+- total_train_batch_size: 128
 - total_eval_batch_size: 6
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine