ConicCat/Qwen2.5-72B-Instruct-abliterated-FP8-Dynamic

Bog standard fp8 w8a8 quant of huihui-ai/Qwen2.5-72B-Instruct-abliterated for datagen purposes.

Recipe:

from transformers import AutoTokenizer, AutoModelForCausalLM

MODEL_ID = "huihui-ai/Qwen2.5-72B-Instruct-abliterated"

model = AutoModelForCausalLM.from_pretrained(MODEL_ID, torch_dtype="auto")
tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)

from llmcompressor import oneshot
from llmcompressor.modifiers.quantization import QuantizationModifier

# Configure the simple PTQ quantization
recipe = QuantizationModifier(
  targets="Linear", scheme="FP8_DYNAMIC", ignore=["lm_head"])

# Apply the quantization algorithm.
oneshot(model=model, recipe=recipe)

# Save the model.
SAVE_DIR = MODEL_ID.rstrip("/").split("/")[-1] + "-FP8-Dynamic"
model.save_pretrained(SAVE_DIR)
tokenizer.save_pretrained(SAVE_DIR)