README.md

metadata

license: mit
datasets:
  - Xilabs/instructmix
  - CreitinGameplays/small-chat-assistant-for-bloom
  - sahil2801/CodeAlpaca-20k
language:
  - en
tags:
  - uncensored
  - unrestricted
  - code
  - biology
  - chemistry
  - finance
  - legal
  - music
  - art
  - climate
  - merge
  - text-generation-inference
  - moe
  - TensorBlock
  - GGUF
widget:
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> who was
      Nikola Tesla? </s> <|assistant|>
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> write a story
      about a cat. </s> <|assistant|>
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> what is an
      essay? </s> <|assistant|>
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> Tell me 5
      Brazilian waterfalls to visit. </s> <|assistant|>
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> write a story
      about how a virus called COVID-19 destroyed the world </s> <|assistant|>
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> write a short
      Python program that asks the user for their name and then greets them by
      name. </s> <|assistant|>
  - text: >-
      <|system|> You are a helpful AI assistant. </s> <|prompter|> What can you
      do? </s> <|assistant|>
inference:
  parameters:
    temperature: 0.1
    do_sample: false
    top_k: 50
    top_p: 0.15
    max_new_tokens: 250
    repetition_penalty: 1.155
base_model: CreitinGameplays/bloom-3b-conversational

Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server

CreitinGameplays/bloom-3b-conversational - GGUF

This repo contains GGUF format model files for CreitinGameplays/bloom-3b-conversational.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4242.

Run them on the TensorBlock client using your local machine ↗

Prompt template

<|system|>{system_prompt}</s><|prompter|>{prompt}</s><|assistant|>

Model file specification

Filename	Quant type	File Size	Description
bloom-3b-conversational-Q2_K.gguf	Q2_K	1.628 GB	smallest, significant quality loss - not recommended for most purposes
bloom-3b-conversational-Q3_K_S.gguf	Q3_K_S	1.833 GB	very small, high quality loss
bloom-3b-conversational-Q3_K_M.gguf	Q3_K_M	2.045 GB	very small, high quality loss
bloom-3b-conversational-Q3_K_L.gguf	Q3_K_L	2.165 GB	small, substantial quality loss
bloom-3b-conversational-Q4_0.gguf	Q4_0	2.232 GB	legacy; small, very high quality loss - prefer using Q3_K_M
bloom-3b-conversational-Q4_K_S.gguf	Q4_K_S	2.242 GB	small, greater quality loss
bloom-3b-conversational-Q4_K_M.gguf	Q4_K_M	2.400 GB	medium, balanced quality - recommended
bloom-3b-conversational-Q5_0.gguf	Q5_0	2.607 GB	legacy; medium, balanced quality - prefer using Q4_K_M
bloom-3b-conversational-Q5_K_S.gguf	Q5_K_S	2.607 GB	large, low quality loss - recommended
bloom-3b-conversational-Q5_K_M.gguf	Q5_K_M	2.734 GB	large, very low quality loss - recommended
bloom-3b-conversational-Q6_K.gguf	Q6_K	3.006 GB	very large, extremely low quality loss
bloom-3b-conversational-Q8_0.gguf	Q8_0	3.888 GB	very large, extremely low quality loss - not recommended

Downloading instruction

Command line

Firstly, install Huggingface Client

pip install -U "huggingface_hub[cli]"

Then, downoad the individual model file the a local directory

huggingface-cli download tensorblock/bloom-3b-conversational-GGUF --include "bloom-3b-conversational-Q2_K.gguf" --local-dir MY_LOCAL_DIR

If you wanna download multiple model files with a pattern (e.g., *Q4_K*gguf), you can try:

huggingface-cli download tensorblock/bloom-3b-conversational-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'