File size: 2,152 Bytes
1ec3373 f4ec97d 1ec3373 f4ec97d 1ec3373 f4ec97d 1ec3373 f4ec97d 1ec3373 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
---
tags:
- autotrain
- text-generation-inference
- text-generation
- peft
library_name: transformers
widget:
- messages:
- role: user
content: What is your favorite condiment?
license: other
datasets:
- timdettmers/openassistant-guanaco
language:
- en
---
# Model Details
This model is a finetuned Meta-Llama-3-8b-Instruct model on the openassistant dataset. It was finetuned using PEFT, a library for efficiently adapting pre-trained language models to various downstream applications without fine-tuning all the model’s parameters.
# Inference with PEFT Models:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel, PeftConfig
base_model = "meta-llama/Meta-Llama-3-8B"
adapter_model = "pantelnm/llama3-openassistant"
prompt = "Write your prompt here!"
model = AutoModelForCausalLM.from_pretrained(base_model)
model = PeftModel.from_pretrained(model, adapter_model)
tokenizer = AutoTokenizer.from_pretrained(base_model)
model = model.to("cuda")
model.eval()
inputs = tokenizer(prompt, return_tensors="pt")
with torch.no_grad():
outputs = model.generate(input_ids=inputs["input_ids"].to("cuda"), max_new_tokens=10)
print(tokenizer.batch_decode(outputs.detach().cpu().numpy(), skip_special_tokens=True)[0])
```
# Model Trained Using AutoTrain
This model was trained using AutoTrain. For more information, please visit [AutoTrain](https://hf.co/docs/autotrain).
# General Usage
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_path = "PATH_TO_THIS_REPO"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(
model_path,
device_map="auto",
torch_dtype='auto'
).eval()
# Prompt content: "hi"
messages = [
{"role": "user", "content": "hi"}
]
input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
output_ids = model.generate(input_ids.to('cuda'))
response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
# Model response: "Hello! How can I assist you today?"
print(response)
``` |