CodeGemma-2B-dora / README.md
abideen's picture
Update README.md
331cad0 verified
|
raw
history blame
2.15 kB
metadata
license: cc-by-nc-4.0
base_model: johnsnowlabs/CodeGemma-2B-Slerp
tags:
  - generated_from_trainer
  - instruct
  - finetune
  - chatml
  - gpt4
  - synthetic data
  - distillation
model-index:
  - name: CodeGemma-2B-Slerp-dora
    results: []
datasets:
  - argilla/distilabel-intel-orca-dpo-pairs
language:
  - en
library_name: transformers
pipeline_tag: text-generation

CodeGemma-2B-Slerp-dora

image/png

CodeGemma-2B-Slerp-dora is a DPO fine-tuned of johnsnowlabs/CodeGemma-2B-Slerp on argilla/distilabel-intel-orca-dpo-pairs preference dataset using DoRA. The model has been trained for 1080 steps. All hyperparams are given below.

πŸ† Evaluation results

Coming Soom

Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "johnsnowlabs/CodeGemma-2B-dora"
messages = [{"role": "user", "content": "Explain what is Machine learning."}]

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
  • optimizer: PagedAdamW with 32-bit precision
  • lr_scheduler_type: Cosine
  • lr_scheduler_warmup_steps: 100
  • training_steps: 1080

LoRA Config

  • lora_r: 16
  • lora_alpha: 32
  • lora_dropout: 0.05
  • peft_use_dora: true

Framework versions

  • Transformers 4.39.0.dev0
  • Peft 0.9.1.dev0
  • Datasets 2.18.0
  • torch 2.2.0
  • accelerate 0.27.2