metadata

license: cc-by-nc-4.0
base_model: PhigRange-2.7B-slerp
tags:
  - generated_from_trainer
  - DPO
  - instruct
  - finetune
  - chatml
  - gpt4
  - synthetic data
  - distillation
model-index:
  - name: PhigRange-DPO
    results: []
datasets:
  - mlabonne/chatml-OpenHermes2.5-dpo-binarized-alpha
language:
  - en
library_name: transformers
pipeline_tag: text-generation

PhigRange-DPO

PhigRange-DPO is a DPO fine-tuned of johnsnowlabs/PhigRange-2.7B-Slerp using the mlabonne/chatml-OpenHermes2.5-dpo-binarized-alpha preference dataset. The model has been trained for for 1080 steps.

🏆 Evaluation results

Coming Soon

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "johnsnowlabs/PhigRange-DPO"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-04
train_batch_size: 1
gradient_accumulation_steps: 8
total_train_batch_size: 8
optimizer: AdamOptimizer32bit
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
training_steps: 1080

Framework versions

Transformers 4.38.0.dev0
Pytorch 2.1.2+cu118
Datasets 2.17.0
Tokenizers 0.15.0