Quantization made by Richard Erkhov.

Llama-2-7b-ultrachat200k - GGUF

Model creator: https://huggingface.co/neuralmagic/
Original model: https://huggingface.co/neuralmagic/Llama-2-7b-ultrachat200k/

Name	Quant method	Size
Llama-2-7b-ultrachat200k.Q2_K.gguf	Q2_K	2.36GB
Llama-2-7b-ultrachat200k.IQ3_XS.gguf	IQ3_XS	2.6GB
Llama-2-7b-ultrachat200k.IQ3_S.gguf	IQ3_S	2.75GB
Llama-2-7b-ultrachat200k.Q3_K_S.gguf	Q3_K_S	2.75GB
Llama-2-7b-ultrachat200k.IQ3_M.gguf	IQ3_M	2.9GB
Llama-2-7b-ultrachat200k.Q3_K.gguf	Q3_K	3.07GB
Llama-2-7b-ultrachat200k.Q3_K_M.gguf	Q3_K_M	3.07GB
Llama-2-7b-ultrachat200k.Q3_K_L.gguf	Q3_K_L	3.35GB
Llama-2-7b-ultrachat200k.IQ4_XS.gguf	IQ4_XS	3.4GB
Llama-2-7b-ultrachat200k.Q4_0.gguf	Q4_0	3.56GB
Llama-2-7b-ultrachat200k.IQ4_NL.gguf	IQ4_NL	3.58GB
Llama-2-7b-ultrachat200k.Q4_K_S.gguf	Q4_K_S	3.59GB
Llama-2-7b-ultrachat200k.Q4_K.gguf	Q4_K	3.8GB
Llama-2-7b-ultrachat200k.Q4_K_M.gguf	Q4_K_M	3.8GB
Llama-2-7b-ultrachat200k.Q4_1.gguf	Q4_1	3.95GB
Llama-2-7b-ultrachat200k.Q5_0.gguf	Q5_0	4.33GB
Llama-2-7b-ultrachat200k.Q5_K_S.gguf	Q5_K_S	4.33GB
Llama-2-7b-ultrachat200k.Q5_K.gguf	Q5_K	4.45GB
Llama-2-7b-ultrachat200k.Q5_K_M.gguf	Q5_K_M	4.45GB
Llama-2-7b-ultrachat200k.Q5_1.gguf	Q5_1	4.72GB
Llama-2-7b-ultrachat200k.Q6_K.gguf	Q6_K	5.15GB
Llama-2-7b-ultrachat200k.Q8_0.gguf	Q8_0	6.67GB

Original model description:

base_model: meta-llama/Llama-2-7b-hf inference: true model_type: llama pipeline_tag: text-generation datasets: - HuggingFaceH4/ultrachat_200k tags: - chat

Llama-2-7b-ultrachat

This repo contains a Llama 2 7B finetuned for chat tasks using the UltraChat 200k dataset.

Official model weights from Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment.

Authors: Neural Magic, Cerebras

Usage

Below we share some code snippets on how to get quickly started with running the model.

Sparse Transfer

By leveraging a pre-sparsified model's structure, you can efficiently fine-tune on new data, leading to reduced hyperparameter tuning, training times, and computational costs. Learn about this process here.

Running the model

This model may be run with the transformers library. For accelerated inference with sparsity, deploy with nm-vllm or deepsparse.

# pip install transformers accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("neuralmagic/Llama-2-7b-ultrachat")
model = AutoModelForCausalLM.from_pretrained("neuralmagic/Llama-2-7b-ultrachat", device_map="auto")

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer.apply_chat_template(input_text, add_generation_prompt=True, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

Evaluation Benchmark Results

Model evaluation metrics and results.

Benchmark	Metric	Llama-2-7b-ultrachat	Llama-2-7b-pruned50-retrained-ultrachat
MMLU	5-shot, top-1	xxxx	xxxx
HellaSwag	0-shot	xxxx	xxxx
WinoGrande	partial score	xxxx	xxxx
ARC-c		xxxx	xxxx
TruthfulQA	5-shot	xxxx	xxxx
HumanEval	pass@1	xxxx	xxxx
GSM8K	maj@1	xxxx	xxxx

Model Training Details

Coming soon.

Help

For further support, and discussions on these models and AI in general, join Neural Magic's Slack Community