Downloads · 30 days
116
0% of all-time downloads
Qwen2.5-72B-Instruct-abliterated-FP8-Dynamic
ConicCat/Qwen2.5-72B-Instruct-abliterated-FP8-Dynamic
Qwen2.5-72B-Instruct-abliterated-FP8-Dynamic is a machine learning model from ConicCat. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Bog standard fp8 w8a8 quant of huihui-ai/Qwen2.5-72B-Instruct-abliterated for datagen purposes.
All-time downloads
108K
Public
Parameters
72.7B
75.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors75.2 GB · 100%
Parameter types
How the weights are stored.
F8_E4M370.2B · 97%
Base models
Model card
From the Hugging Face model README
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)
- Type
- qwen2
- Created
- Oct 15, 2025
- Updated
- Oct 15, 2025