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

Llama 3 15B Instruct Zeroed Ft by elinas

 »  All LLMs  »  elinas  »  Llama 3 15B Instruct Zeroed Ft   URL Share it on

  97   
  Autotrain compatible Base model:elinas/llama-3-15b-... Base model:finetune:elinas/lla...   Conversational Dataset:chat-error/pure-dove-s...   Endpoints compatible   Instruct   Llama   Merge   Mergekit   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/elinas/Llama-3-15B-Instruct-zeroed-ft 

Llama 3 15B Instruct Zeroed Ft Benchmarks

ARC: 60.75 vs 96.7 (so35)-37.2%
HellaSwag: 78.88 vs 95.3 (gpt4)-17.2%
MMLU: 67.71 vs 88.3 (so35)-23.3%
TruthfulQA: 52.45 vs 59 (gpt4)-11.1%
WinoGrande: 76.24 vs 87.5 (gpt4)-12.9%
GSM8K: 71.57 vs 96.4 (so35)-25.8%
LLME Score: 0.21499
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3 15B Instruct Zeroed Ft (elinas/Llama-3-15B-Instruct-zeroed-ft)

Llama 3 15B Instruct Zeroed Ft Parameters and Internals

Model Type 
text generation, language model
Additional Notes 
Experiment aimed at stabilizing the model post fine-tuning with potential for continued fine-tuning.
Training Details 
Data Sources:
Chat-Error/Pure-dove-sharegpt
Methodology:
QLoRA finetuning on a zeroed passthrough merge of pre-trained language models.
Context Length:
8192
Training Time:
2h 30m
Hardware Used:
3x3090 GPUs
LLM NameLlama 3 15B Instruct Zeroed Ft
Repository 🤗https://huggingface.co/elinas/Llama-3-15B-Instruct-zeroed-ft 
Base Model(s)  Llama 3 15B Instruct Zeroed   elinas/Llama-3-15B-Instruct-zeroed
Model Size15b
Required VRAM30.1 GB
Updated2024-12-21
Maintainerelinas
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-7   5.0 GB: 2-of-7   4.9 GB: 3-of-7   5.0 GB: 4-of-7   4.9 GB: 5-of-7   4.2 GB: 6-of-7   1.1 GB: 7-of-7
Model ArchitectureLlamaForCausalLM
Licensellama3
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 Llama 3 15B Instruct Zeroed Ft

Best Alternatives
Context / RAM
Downloads
Likes
L3 Aethora 15B V28K / 30.1 GB11040
OpenCrystal 15B L3 V38K / 30 GB1759
OpenCrystal 15B L3 V28K / 30 GB275
OpenCrystal L3 15B V2.18K / 30 GB115
Meta Llama 3 15B Instruct8K / 23.9 GB101
Llama 3 15B Instruct Ft V28K / 30.1 GB944
LLaMa 3 Stheno V3.2 15B8K / 30 GB185
LLaMa 3 Stheno V3.2 15B8K / 30 GB175
L3 NA Aethora 15B8K / 30.1 GB163
L3 NA Aethora 15B8K / 30.1 GB63
Note: green Score (e.g. "73.2") means that the model is better than elinas/Llama-3-15B-Instruct-zeroed-ft.

Rank the Llama 3 15B Instruct Zeroed Ft 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? 40013 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 v20241217
  Support LLM Explorer