Quantization made by Richard Erkhov.

megatron_v1 - GGUF

Model creator: https://huggingface.co/Eurdem/
Original model: https://huggingface.co/Eurdem/megatron_v1/

Name	Quant method	Size
megatron_v1.Q2_K.gguf	Q2_K	4.43GB
megatron_v1.IQ3_XS.gguf	IQ3_XS	4.95GB
megatron_v1.IQ3_S.gguf	IQ3_S	5.22GB
megatron_v1.Q3_K_S.gguf	Q3_K_S	5.2GB
megatron_v1.IQ3_M.gguf	IQ3_M	5.35GB
megatron_v1.Q3_K.gguf	Q3_K	5.78GB
megatron_v1.Q3_K_M.gguf	Q3_K_M	5.78GB
megatron_v1.Q3_K_L.gguf	Q3_K_L	6.27GB
megatron_v1.IQ4_XS.gguf	IQ4_XS	6.5GB
megatron_v1.Q4_0.gguf	Q4_0	6.78GB
megatron_v1.IQ4_NL.gguf	IQ4_NL	6.85GB
megatron_v1.Q4_K_S.gguf	Q4_K_S	6.84GB
megatron_v1.Q4_K.gguf	Q4_K	7.25GB
megatron_v1.Q4_K_M.gguf	Q4_K_M	7.25GB
megatron_v1.Q4_1.gguf	Q4_1	7.52GB
megatron_v1.Q5_0.gguf	Q5_0	8.26GB
megatron_v1.Q5_K_S.gguf	Q5_K_S	8.26GB
megatron_v1.Q5_K.gguf	Q5_K	8.51GB
megatron_v1.Q5_K_M.gguf	Q5_K_M	8.51GB
megatron_v1.Q5_1.gguf	Q5_1	9.01GB
megatron_v1.Q6_K.gguf	Q6_K	9.84GB
megatron_v1.Q8_0.gguf	Q8_0	12.75GB

Original model description:

license: apache-2.0 tags: - moe - merge model-index: - name: megatron_v1 results: - task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics: - type: acc_norm value: 65.96 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Eurdem/megatron_v1 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics: - type: acc_norm value: 84.8 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Eurdem/megatron_v1 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics: - type: acc value: 65.02 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Eurdem/megatron_v1 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics: - type: mc2 value: 60.32 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Eurdem/megatron_v1 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics: - type: acc value: 79.79 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Eurdem/megatron_v1 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 57.01 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Eurdem/megatron_v1 name: Open LLM Leaderboard

megatron_v1

megatron_v1 is a Mixure of Experts (MoE) made of mistral models.

💻 Usage

from transformers import AutoTokenizer
import transformers
import torch

model = "Eurdem/megatron_v1"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)

messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
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"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.82
AI2 Reasoning Challenge (25-Shot)	65.96
HellaSwag (10-Shot)	84.80
MMLU (5-Shot)	65.02
TruthfulQA (0-shot)	60.32
Winogrande (5-shot)	79.79
GSM8k (5-shot)	57.01