LLM EXPLORER 56,604 MODELS INDEXED

Megatron GPT 2 345M EvolInstruct by KnutJaegersberg

By KnutJaegersberg · 174 downloads
HF Repo ↗ Copy URL

Megatron GPT 2 345M EvolInstruct is an open-source language model by KnutJaegersberg. Features: 345m LLM, VRAM: 1.4GB, License: cc-by-nc-4.0, LLM Explorer Score: 0.12, Arc: 24.1, HellaSwag: 35.1, MMLU: 24.5, GSM8K: 0.4.

  Endpoints compatible   Gpt2   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/KnutJaegersberg/megatron-GPT-2-345m-EvolInstruct 

Megatron GPT 2 345M EvolInstruct Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
Research
Applications:
Benchmarking, Experimental setups
Primary Use Cases:
Text generation tasks
Limitations:
Limited to non-commercial use
Additional Notes 
Benchmark information also includes results on specific open LLM leaderboard metrics.
Supported Languages 
English (fluent)
Training Details 
Data Sources:
Evol-Instruct dataset from WizardLM
Methodology:
Fine-tuning
Model Architecture:
Transformer-based architecture
Input Output 
Input Format:
Text prompts
Accepted Modalities:
text
Output Format:
Generated text
Performance Tips:
Utilize GPUs for optimal performance.
LLM NameMegatron GPT 2 345M EvolInstruct
Repository 🤗https://huggingface.co/KnutJaegersberg/megatron-GPT-2-345m-EvolInstruct 
Model Size345m
Required VRAM1.4 GB
Updated2026-07-12
MaintainerKnutJaegersberg
Model Typegpt2
Model Files  1.4 GB   1.4 GB   0.0 GB
Model ArchitectureGPT2LMHeadModel
Licensecc-by-nc-4.0
Transformers Version4.29.0.dev0
Vocabulary Size50257
Torch Data Typefloat32
Activation Functiongelu_new

Best Alternatives to Megatron GPT 2 345M EvolInstruct

Best Alternatives
Context / RAM
Downloads
Likes
Anwgpt2 345M0K / 1.4 GB460
...ron Gpt2 345M Evol Instruct V20K / 1.4 GB1633
Megatron Gpt2 345M0K / 0.7 GB230279
Note: green Score (e.g. "73.2") means that the model is better than KnutJaegersberg/megatron-GPT-2-345m-EvolInstruct.