HKT-vul-DeepSeek-R1-8b-it-v0.2

HKT-vul-DeepSeek-R1-8b-it-v0.2

This is a LoRA fine-tuned version of unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit.

Training Details

• Base Model: unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit
• Fine-tuning Method: LoRA
• Merge Method: merge_and_unload()

Usage

Install necessary libraries

import os
if "COLAB_" not in "".join(os.environ.keys()):
    !pip install unsloth
else:
    # Do this only in Colab notebooks! Otherwise use pip install unsloth
    !pip install --no-deps bitsandbytes accelerate xformers==0.0.29.post3 peft trl==0.15.2 triton cut_cross_entropy unsloth_zoo
    !pip install sentencepiece protobuf datasets huggingface_hub hf_transfer
    !pip install --no-deps unsloth

Install model

from unsloth import FastLanguageModel
import torch
max_seq_length = 5000 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.

model_name = "weifar/unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit"

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = model_name,
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
    # token = "hf_...", # use one if using gated models like meta-llama/Llama-2-7b-hf
)

model = FastLanguageModel.get_peft_model(
    model,
    r = 16, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128
    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
                      "gate_proj", "up_proj", "down_proj",],
    lora_alpha = 16,
    lora_dropout = 0, # Supports any, but = 0 is optimized
    bias = "none",    # Supports any, but = "none" is optimized
    # [NEW] "unsloth" uses 30% less VRAM, fits 2x larger batch sizes!
    use_gradient_checkpointing = "unsloth", # True or "unsloth" for very long context
    random_state = 3407,
    use_rslora = False,  # We support rank stabilized LoRA
    loftq_config = None, # And LoftQ
)

Use model

FastLanguageModel.for_inference(model) # Enable native 2x faster inference

inputs = tokenizer(eval_prompt, return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 4000)