apepkuss79's picture
Update README.md
cfb7120 verified
metadata
base_model: berkeley-nest/Starling-LM-7B-alpha
datasets:
  - berkeley-nest/Nectar
inference: false
library_name: transformers
license: cc-by-nc-4.0
model_creator: Berkeley-Nest
model_name: Starling LM 7B Alpha
model_type: mistral
pipeline_tag: text-generation
quantized_by: Second State Inc.
language:
  - en
tags:
  - reward model
  - RLHF
  - RLAIF

Starling-LM-7B-alpha-GGUF

Original Model

berkeley-nest/Starling-LM-7B-alpha

Run with LlamaEdge

  • LlamaEdge version: v0.2.8 and above

  • Prompt template

    • Prompt type: openchat

    • Prompt string

      GPT4 User: {prompt}<|end_of_turn|>GPT4 Assistant:
      
    • Reverse prompt: <|end_of_turn|>

  • Context size: 4096

  • Run as LlamaEdge service

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:starling-lm-7b-alpha.Q5_K_M.gguf llama-api-server.wasm -p openchat -r '<|end_of_turn|>'
    
  • Run as LlamaEdge command app

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:starling-lm-7b-alpha.Q5_K_M.gguf llama-chat.wasm -p openchat -r '<|end_of_turn|>'
    

Quantized GGUF Models

Name Quant method Bits Size Use case
Starling-LM-7B-alpha-Q2_K.gguf Q2_K 2 2.7 GB smallest, significant quality loss - not recommended for most purposes
Starling-LM-7B-alpha-Q3_K_L.gguf Q3_K_L 3 3.82 GB small, substantial quality loss
Starling-LM-7B-alpha-Q3_K_M.gguf Q3_K_M 3 3.52 GB very small, high quality loss
Starling-LM-7B-alpha-Q3_K_S.gguf Q3_K_S 3 3.16 GB very small, high quality loss
Starling-LM-7B-alpha-Q4_0.gguf Q4_0 4 4.11 GB legacy; small, very high quality loss - prefer using Q3_K_M
Starling-LM-7B-alpha-Q4_K_M.gguf Q4_K_M 4 4.37 GB medium, balanced quality - recommended
Starling-LM-7B-alpha-Q4_K_S.gguf Q4_K_S 4 4.14 GB small, greater quality loss
Starling-LM-7B-alpha-Q5_0.gguf Q5_0 5 5.00 GB legacy; medium, balanced quality - prefer using Q4_K_M
Starling-LM-7B-alpha-Q5_K_M.gguf Q5_K_M 5 5.13 GB large, very low quality loss - recommended
Starling-LM-7B-alpha-Q5_K_S.gguf Q5_K_S 5 5.00 GB large, low quality loss - recommended
Starling-LM-7B-alpha-Q6_K.gguf Q6_K 6 5.94 GB very large, extremely low quality loss
Starling-LM-7B-alpha-Q8_0.gguf Q8_0 8 7.70 GB very large, extremely low quality loss - not recommended