RichardErkhov/DreadPoor_-_Harpy-7B-Model_Stock-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Harpy-7B-Model_Stock - GGUF

• Model creator: https://huggingface.co/DreadPoor/
• Original model: https://huggingface.co/DreadPoor/Harpy-7B-Model_Stock/

Name	Quant method	Size
Harpy-7B-Model_Stock.Q2_K.gguf	Q2_K	2.53GB
Harpy-7B-Model_Stock.IQ3_XS.gguf	IQ3_XS	2.81GB
Harpy-7B-Model_Stock.IQ3_S.gguf	IQ3_S	2.96GB
Harpy-7B-Model_Stock.Q3_K_S.gguf	Q3_K_S	2.95GB
Harpy-7B-Model_Stock.IQ3_M.gguf	IQ3_M	3.06GB
Harpy-7B-Model_Stock.Q3_K.gguf	Q3_K	3.28GB
Harpy-7B-Model_Stock.Q3_K_M.gguf	Q3_K_M	3.28GB
Harpy-7B-Model_Stock.Q3_K_L.gguf	Q3_K_L	3.56GB
Harpy-7B-Model_Stock.IQ4_XS.gguf	IQ4_XS	3.67GB
Harpy-7B-Model_Stock.Q4_0.gguf	Q4_0	3.83GB
Harpy-7B-Model_Stock.IQ4_NL.gguf	IQ4_NL	3.87GB
Harpy-7B-Model_Stock.Q4_K_S.gguf	Q4_K_S	3.86GB
Harpy-7B-Model_Stock.Q4_K.gguf	Q4_K	4.07GB
Harpy-7B-Model_Stock.Q4_K_M.gguf	Q4_K_M	4.07GB
Harpy-7B-Model_Stock.Q4_1.gguf	Q4_1	4.24GB
Harpy-7B-Model_Stock.Q5_0.gguf	Q5_0	4.65GB
Harpy-7B-Model_Stock.Q5_K_S.gguf	Q5_K_S	4.65GB
Harpy-7B-Model_Stock.Q5_K.gguf	Q5_K	4.78GB
Harpy-7B-Model_Stock.Q5_K_M.gguf	Q5_K_M	4.78GB
Harpy-7B-Model_Stock.Q5_1.gguf	Q5_1	5.07GB
Harpy-7B-Model_Stock.Q6_K.gguf	Q6_K	5.53GB
Harpy-7B-Model_Stock.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0 tags: - merge - mergekit - lazymergekit model-index: - name: Harpy-7B-Model_Stock 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: 73.21 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=DreadPoor/Harpy-7B-Model_Stock 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: 88.72 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=DreadPoor/Harpy-7B-Model_Stock 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: 65.07 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=DreadPoor/Harpy-7B-Model_Stock 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: 71.35 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=DreadPoor/Harpy-7B-Model_Stock 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: 85.24 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=DreadPoor/Harpy-7B-Model_Stock 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: 69.45 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=DreadPoor/Harpy-7B-Model_Stock name: Open LLM Leaderboard

Harpy-7B-Model_Stock

Harpy-7B-Model_Stock is a merge of the following models using LazyMergekit:

🧩 Configuration

models:
  - model: Endevor/InfinityRP-v1-7B
  - model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
  - model: abideen/AlphaMonarch-laser
merge_method: model_stock
base_model: Endevor/InfinityRP-v1-7B
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "DreadPoor/Harpy-7B-Model_Stock"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

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.	75.51
AI2 Reasoning Challenge (25-Shot)	73.21
HellaSwag (10-Shot)	88.72
MMLU (5-Shot)	65.07
TruthfulQA (0-shot)	71.35
Winogrande (5-shot)	85.24
GSM8k (5-shot)	69.45