RichardErkhov/AIGym_-_deepseek-coder-1.3b-chat-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

deepseek-coder-1.3b-chat - GGUF

• Model creator: https://huggingface.co/AIGym/
• Original model: https://huggingface.co/AIGym/deepseek-coder-1.3b-chat/

Name	Quant method	Size
deepseek-coder-1.3b-chat.Q2_K.gguf	Q2_K	0.52GB
deepseek-coder-1.3b-chat.IQ3_XS.gguf	IQ3_XS	0.57GB
deepseek-coder-1.3b-chat.IQ3_S.gguf	IQ3_S	0.6GB
deepseek-coder-1.3b-chat.Q3_K_S.gguf	Q3_K_S	0.6GB
deepseek-coder-1.3b-chat.IQ3_M.gguf	IQ3_M	0.63GB
deepseek-coder-1.3b-chat.Q3_K.gguf	Q3_K	0.66GB
deepseek-coder-1.3b-chat.Q3_K_M.gguf	Q3_K_M	0.66GB
deepseek-coder-1.3b-chat.Q3_K_L.gguf	Q3_K_L	0.69GB
deepseek-coder-1.3b-chat.IQ4_XS.gguf	IQ4_XS	0.7GB
deepseek-coder-1.3b-chat.Q4_0.gguf	Q4_0	0.72GB
deepseek-coder-1.3b-chat.IQ4_NL.gguf	IQ4_NL	0.73GB
deepseek-coder-1.3b-chat.Q4_K_S.gguf	Q4_K_S	0.76GB
deepseek-coder-1.3b-chat.Q4_K.gguf	Q4_K	0.81GB
deepseek-coder-1.3b-chat.Q4_K_M.gguf	Q4_K_M	0.81GB
deepseek-coder-1.3b-chat.Q4_1.gguf	Q4_1	0.8GB
deepseek-coder-1.3b-chat.Q5_0.gguf	Q5_0	0.87GB
deepseek-coder-1.3b-chat.Q5_K_S.gguf	Q5_K_S	0.89GB
deepseek-coder-1.3b-chat.Q5_K.gguf	Q5_K	0.93GB
deepseek-coder-1.3b-chat.Q5_K_M.gguf	Q5_K_M	0.93GB
deepseek-coder-1.3b-chat.Q5_1.gguf	Q5_1	0.95GB
deepseek-coder-1.3b-chat.Q6_K.gguf	Q6_K	1.09GB
deepseek-coder-1.3b-chat.Q8_0.gguf	Q8_0	1.33GB

Original model description:

license: apache-2.0 tags: - finetuned pipeline_tag: text-generation model-index: - name: deepseek-coder-1.3b-chat 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: 25.85 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=AIGym/deepseek-coder-1.3b-chat 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: 39.59 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=AIGym/deepseek-coder-1.3b-chat 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: 26.36 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=AIGym/deepseek-coder-1.3b-chat 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: 43.92 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=AIGym/deepseek-coder-1.3b-chat 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: 51.7 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=AIGym/deepseek-coder-1.3b-chat 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: 3.03 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=AIGym/deepseek-coder-1.3b-chat name: Open LLM Leaderboard

deepseek-coder-1.3b-chat

It was created by starting with the deepseek-coder-1.3b and training it on the open assistant dataset. We have attached the wandb report in pdf form to view the training run at a glance.

Reson

This model was fine tned to allow it to follow direction and is a steeping stone to further training, but still would be good for asking qestions about code.

How to use

You will need the transformers>=4.31

from transformers import AutoTokenizer
import transformers 
import torch
model = "AIGym/deepseek-coder-1.3b-chat"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

prompt = "What are the values in open source projects?"
formatted_prompt = (
    f"### Human: {prompt}### Assistant:"
)


sequences = pipeline(
    formatted_prompt,
    do_sample=True,
    top_k=50,
    top_p = 0.7,
    num_return_sequences=1,
    repetition_penalty=1.1,
    max_new_tokens=500,
)
for seq in sequences:
    print(f"Result: {seq['generated_text']}")

Referrals

Run Pod - This is who I use to train th emodels on huggingface. If you use it we both get free crdits. - Visit Runpod's Website!

Paypal - If you want to leave a tip, it is appecaheted. - Visit My Paypal!

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	31.74
AI2 Reasoning Challenge (25-Shot)	25.85
HellaSwag (10-Shot)	39.59
MMLU (5-Shot)	26.36
TruthfulQA (0-shot)	43.92
Winogrande (5-shot)	51.70
GSM8k (5-shot)	3.03