LLM EXPLORER 59,420 MODELS INDEXED

Gpt2 Final by DrNicefellow

By DrNicefellow · 6 downloads

Gpt2 Final is an open-source language model by DrNicefellow. Features: 124m LLM, VRAM: 0.5GB, License: apache-2.0, LLM Explorer Score: 0.17, Arc: 21.6, HellaSwag: 26, MMLU: 24.8.

  Endpoints compatible   Gpt2   Pytorch   Region:us   Safetensors

Gpt2 Final Parameters and Internals

Additional Notes 
The tokenizer used is from Hugging Face's OpenAI Community GPT-2.
Training Details 
Data Sources:
40B tokens
Data Volume:
40B tokens
Model Architecture:
GPT-2
LLM NameGpt2 Final
Repository πŸ€—https://huggingface.co/DrNicefellow/gpt2-final 
Model Size124m
Required VRAM0.5 GB
Updated2026-08-03
MaintainerDrNicefellow
Model Typegpt2
Model Files  0.5 GB   0.5 GB
Model ArchitectureGPT2LMHeadModel
Licenseapache-2.0
Model Max Length1024
Transformers Version4.38.0.dev0
Tokenizer ClassGPT2Tokenizer
Vocabulary Size50257
Torch Data Typefloat32
Activation Functiongelu_new
Errorsreplace

Best Alternatives to Gpt2 Final

Best Alternatives
Context / RAM
Downloads
Likes
Chris GPT 2 124M0K / 0.5 GB2080
BrtGPT 124M Base0K / 0.4 GB601
GPT2 124M Poetry RL0K / 0.5 GB52
LaMini GPT 124M0K / 0.5 GB1445523
Filiberto 124M0K / 0.5 GB60
GPT2 Nepali 124M0K / 0.5 GB295
GPT2 Nepali 124M0K / 0.5 GB45
Dlite V2 124M0K / 0.3 GB1337
Dlite V1 124M0K / 0.5 GB1310
Note: green Score (e.g. "73.2") means that the model is better than DrNicefellow/gpt2-final.