LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Bloom Zh 3B Chat by ikala

 »  All LLMs  »  ikala  »  Bloom Zh 3B Chat   URL Share it on

  2225   
  Autotrain compatible   Bloom Dataset:anon8231489123/sharegp... Dataset:databricks/databricks-...   Dataset:lium/tedlium   Dataset:openassistant/oasst1 Dataset:theblackcat102/joke ex...   En   Endpoints compatible   Ja   Pytorch   Region:us   Safetensors   Sft   Zh
Model Card on HF 🤗: https://huggingface.co/ikala/bloom-zh-3b-chat 

Bloom Zh 3B Chat Benchmarks

ARC: 38.82 vs 96.7 (so35)-59.9%
HellaSwag: 54.71 vs 95.3 (gpt4)-42.6%
MMLU: 31.62 vs 88.3 (so35)-64.2%
TruthfulQA: 41.25 vs 59 (gpt4)-30.1%
WinoGrande: 58.64 vs 87.5 (gpt4)-33%
GSM8K: 0.45 vs 96.4 (so35)-99.5%
LLME Score: 0.15919
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Bloom Zh 3B Chat (ikala/bloom-zh-3b-chat)

Bloom Zh 3B Chat Parameters and Internals

Model Type 
Transformer-based Language Model
Additional Notes 
It is based on a Bloom-zh's 3B that was fine-tuned on human demonstrations of assistant conversations.
Supported Languages 
en (English), zh (Chinese), ja (Japanese)
Training Details 
Data Sources:
OpenAssistant/oasst1, databricks/databricks-dolly-15k, anon8231489123/ShareGPT_Vicuna_unfiltered, LIUM/tedlium, theblackcat102/joke_explaination
Methodology:
supervised finetune on sequence length of 5120
Hardware Used:
NVIDIA RTX A6000 x 4
Input Output 
Input Format:
<|prompter|> followed by input and ends with <|assistant|> followed by <|assistant|>.
Output Format:
Text generation.
LLM NameBloom Zh 3B Chat
Repository 🤗https://huggingface.co/ikala/bloom-zh-3b-chat 
Model Size3b
Required VRAM6 GB
Updated2025-03-22
Maintainerikala
Model Typebloom
Model Files  6.0 GB   6.0 GB
Supported Languagesen zh ja
Model ArchitectureBloomForCausalLM
Licensebigscience-openrail-m
Transformers Version4.28.0.dev0
Tokenizer ClassBloomTokenizer
Padding Token<pad>
Vocabulary Size250688
Torch Data Typefloat16

Best Alternatives to Bloom Zh 3B Chat

Best Alternatives
Context / RAM
Downloads
Likes
Bloom 3B Conversational0K / 6.1 GB1642
Bloomz 3B Sft Chat0K / 6 GB183012
Blossom V2 3B0K / 6 GB21900
Gpt3 Finnish 3B Instruct0K / 12.7 GB111
Blossom V1 3B0K / 6 GB21881
... Bloom 3B Conversational 4bits0K / 2.6 GB60
Gogpt 3B Bloom0K / 14.6 GB18345
Deer 3B0K / 14.6 GB17592
Bloomz 3B0K / 6 GB3302179
...lish Dental Raw New Pretrained0K / 3.6 GB640
Note: green Score (e.g. "73.2") means that the model is better than ikala/bloom-zh-3b-chat.

Rank the Bloom Zh 3B Chat Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 45389 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer