LLM EXPLORER 60,519 MODELS INDEXED

Reflection 70B Bnb 4bit by unsloth

By unsloth · 85 downloads
HF Repo ↗ Copy URL

Reflection 70B Bnb 4bit is an open-source language model by unsloth. Features: 70b LLM, VRAM: 39.7GB, Context: 128K, License: llama3.1, Quantized, Instruction-Based, LLM Explorer Score: 0.14.

  4-bit   4bit Base model:meta-llama/meta-lla... Base model:quantized:meta-llam...   Bitsandbytes   Conversational   Instruct   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Unsloth
Model Card on HF 🤗: https://huggingface.co/unsloth/Reflection-70B-bnb-4bit 

Reflection 70B Bnb 4bit Parameters and Internals

Model Type 
text-generation
Additional Notes 
The model was trained using a new technique called Reflection-Tuning that teaches it to detect its reasoning mistakes and correct them.
Training Details 
Data Sources:
synthetic data generated by Glaive
Methodology:
Reflection-Tuning
Input Output 
Input Format:
Llama 3.1 chat format
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Recommend using a temperature of 0.7 and a top_p of 0.95. For increased accuracy, append 'Think carefully.' at the end of your messages.
LLM NameReflection 70B Bnb 4bit
Repository 🤗https://huggingface.co/unsloth/Reflection-70B-bnb-4bit 
Base Model(s)  meta-llama/Meta-Llama-3.1-70B-Instruct   meta-llama/Meta-Llama-3.1-70B-Instruct
Model Size70b
Required VRAM39.7 GB
Updated2024-09-06
Maintainerunsloth
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-8   5.0 GB: 2-of-8   5.0 GB: 3-of-8   4.9 GB: 4-of-8   5.0 GB: 5-of-8   5.0 GB: 6-of-8   5.0 GB: 7-of-8   4.8 GB: 8-of-8
Quantization Type4bit
Model ArchitectureLlamaForCausalLM
Licensellama3.1
Context Length131072
Model Max Length131072
Transformers Version4.44.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|finetune_right_pad_id|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Reflection 70B Bnb 4bit

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 75 GB415
...B Instruct Gradient 1048K 4bit1024K / 39.7 GB253
...B Instruct Gradient 1048K 2bit1024K / 21.9 GB252
...0B Instruct Gradient 262K 4bit256K / 39.7 GB253
...0B Instruct Gradient 262K 8bit256K / 75 GB262
...0B Instruct Gradient 262K 2bit256K / 21.9 GB201
... Gradient 262K 2.25bpw H6 EXL2256K / 22.2 GB80
...t Gradient 262K 4.0bpw H6 EXL2256K / 37.2 GB71
...t Gradient 262K 2.4bpw H6 EXL2256K / 23.5 GB60
...t Gradient 262K 3.5bpw H6 EXL2256K / 32.9 GB60