ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2 by OwenArli

 ยป  All LLMs  ยป  OwenArli  ยป  ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2   URL Share it on

  8-bit   Autotrain compatible   Conversational   Endpoints compatible   Exl2   Instruct   Llama   Quantized   Region:us   Safetensors

ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2 (OwenArli/ArliAI-Llama-3-8B-Instruct-Dolfin-v0.1-exl2-h8-8bpw-exl2)

ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2 Parameters and Internals

Model Type 
LlamaForCausalLM
Additional Notes 
The model is intended to be less censored and capable at general tasks. It uses the instruct format for training rather than ChatML for better performance. The training utilized 4-bit loading and Qlora with 64-rank and 128-alpha.
Training Details 
Data Sources:
https://huggingface.co/datasets/cognitivecomputations/dolphin
Methodology:
Trained on a modified and improved version of the Cognitive Computations Eric Hartford's Dolphin dataset. Used Meta's new Llama 3 instruct format.
Context Length:
2048
Training Time:
2 days
Hardware Used:
2x RTX3090
Input Output 
Input Format:
<|begin_of_text|><|start_header_id|>system<|end_header_id|> {{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|> {{ user_message }}<|eot_id|><|start_header_id|>assistant<|end_header_id>
Output Format:
<|eot_id|><|start_header_id|>user<|end_header_id|>
LLM NameArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2
Repository ๐Ÿค—https://huggingface.co/OwenArli/ArliAI-Llama-3-8B-Instruct-Dolfin-v0.1-exl2-h8-8bpw-exl2 
Model Size8b
Required VRAM8.5 GB
Updated2025-01-23
MaintainerOwenArli
Model Typellama
Instruction-BasedYes
Model Files  8.5 GB
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length8192
Model Max Length8192
Transformers Version4.40.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|end_of_text|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 4bit1024K / 4.5 GB312
...B Instruct Gradient 1048K 8bit1024K / 8.6 GB251
...truct Gradient 1048K Bpw6 EXL21024K / 6.7 GB92
...truct Gradient 1048K Bpw5 EXL21024K / 5.8 GB90
Llama 3 8B Instruct 1048K 4bit1024K / 4.5 GB825
Llama 3 8B Instruct 1048K 8bit1024K / 8.6 GB2217
... Gradient 1048K 8.0bpw H8 EXL21024K / 8.6 GB83
...ct Gradient 1048K Bpw2.25 EXL21024K / 3.4 GB81
...B Instruct 262k V2 EXL2 8.0bpw256K / 8.4 GB300
Llama 3 8B Instruct 262K 2bit256K / 2.5 GB171
Note: green Score (e.g. "73.2") means that the model is better than OwenArli/ArliAI-Llama-3-8B-Instruct-Dolfin-v0.1-exl2-h8-8bpw-exl2.

Rank the ArliAI Llama 3 8B Instruct Dolfin V0.1 EXL2 H8 8bpw EXL2 Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 41774 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227