Push model using huggingface_hub.

README.md

@@ -26,7 +26,7 @@ You can then generate text as follows:
 ```python
 from transformers import pipeline
 
-generator = pipeline("text-generation", model="yuansui//tmp/tmp2wrg17ux/yuansui/Meta-Llama-3.1-8B-Instruct-PPO-tuned")
+generator = pipeline("text-generation", model="yuansui//tmp/tmp_w2uh6fd/yuansui/Meta-Llama-3.1-8B-Instruct-PPO-tuned")
 outputs = generator("Hello, my llama is cute")
 ```
 
@@ -36,8 +36,8 @@ If you want to use the model for training or to obtain the outputs from the valu
 from transformers import AutoTokenizer
 from trl import AutoModelForCausalLMWithValueHead
 
-tokenizer = AutoTokenizer.from_pretrained("yuansui//tmp/tmp2wrg17ux/yuansui/Meta-Llama-3.1-8B-Instruct-PPO-tuned")
-model = AutoModelForCausalLMWithValueHead.from_pretrained("yuansui//tmp/tmp2wrg17ux/yuansui/Meta-Llama-3.1-8B-Instruct-PPO-tuned")
+tokenizer = AutoTokenizer.from_pretrained("yuansui//tmp/tmp_w2uh6fd/yuansui/Meta-Llama-3.1-8B-Instruct-PPO-tuned")
+model = AutoModelForCausalLMWithValueHead.from_pretrained("yuansui//tmp/tmp_w2uh6fd/yuansui/Meta-Llama-3.1-8B-Instruct-PPO-tuned")
 
 inputs = tokenizer("Hello, my llama is cute", return_tensors="pt")
 outputs = model(**inputs, labels=inputs["input_ids"])

adapter_config.json

@@ -21,11 +21,11 @@
   "revision": null,
   "target_modules": [
     "up_proj",
-    "k_proj",
-    "v_proj",
-    "o_proj",
     "gate_proj",
+    "v_proj",
     "q_proj",
+    "o_proj",
+    "k_proj",
     "down_proj"
   ],
   "task_type": "CAUSAL_LM",