readme

README.md

@@ -1,16 +1,16 @@
 ---
 license: apache-2.0
 ---
-### Huggingface RWKV Finch 14B Model
+### Huggingface RWKV Finch 7B Model
 
-> HF compatible model for Finch-14B.
+> HF compatible model for Finch-7B.
 
 ![Finch Bird](./imgs/finch.jpg)
 
 
 > **! Important Note !**
 >
-> The following is the HF transformers implementation of the Finch 14B model. This is meant to be used with the huggingface transformers
+> The following is the HF transformers implementation of the Finch 7B model. This is meant to be used with the huggingface transformers
 >
 >
 
@@ -18,15 +18,15 @@ license: apache-2.0
 ## Quickstart with the hugging face transformer library
 
 ```
-model = AutoModelForCausalLM.from_pretrained("RWKV/v6-Finch-14B-HF", trust_remote_code=True).to(torch.float32)
-tokenizer = AutoTokenizer.from_pretrained("RWKV/v6-Finch-14B-HF", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True).to(torch.float32)
+tokenizer = AutoTokenizer.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True)
 ```
 
 ## Evaluation
 
 The following demonstrates the improvements from Eagle 7B to Finch 14B
 
-|  | [Eagle 7B](https://huggingface.co/RWKV/v5-Eagle-7B-HF) | [Finch 7B](https://huggingface.co/RWKV/v6-Finch-7B-HF) | [Finch 14B](https://huggingface.co/RWKV/v6-Finch-14B-HF) |
+|  | [Eagle 7B](https://huggingface.co/RWKV/v6-Finch-7B-HF) | [Finch 7B](https://huggingface.co/RWKV/v6-Finch-7B-HF) | [Finch 14B](https://huggingface.co/RWKV/v6-Finch-14B-HF) |
 | --- | --- | --- | --- |
 | [ARC](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/arc) | 39.59% | 41.47% | 46.33% |
 | [HellaSwag](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/hellaswag) | 53.09% | 55.96% | 57.69% |
@@ -59,8 +59,8 @@ User: {instruction}
 Assistant:"""
 
 
-model = AutoModelForCausalLM.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True).to(torch.float32)
-tokenizer = AutoTokenizer.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True).to(torch.float32)
+tokenizer = AutoTokenizer.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True)
 
 text = "请介绍北京的旅游景点"
 prompt = generate_prompt(text)
@@ -115,8 +115,8 @@ User: {instruction}
 Assistant:"""
 
 
-model = AutoModelForCausalLM.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True, torch_dtype=torch.float16).to(0)
-tokenizer = AutoTokenizer.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True, torch_dtype=torch.float16).to(0)
+tokenizer = AutoTokenizer.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True)
 
 text = "介绍一下大熊猫"
 prompt = generate_prompt(text)
@@ -162,8 +162,8 @@ User: {instruction}
 
 Assistant:"""
 
-model = AutoModelForCausalLM.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True).to(torch.float32)
-tokenizer = AutoTokenizer.from_pretrained("RWKV/v5-Eagle-7B-HF", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True).to(torch.float32)
+tokenizer = AutoTokenizer.from_pretrained("RWKV/v6-Finch-7B-HF", trust_remote_code=True)
 
 texts = ["请介绍北京的旅游景点", "介绍一下大熊猫", "乌兰察布"]
 prompts = [generate_prompt(text) for text in texts]