LLM EXPLORER 58,297 MODELS INDEXED

Griffin C3t 8L V0.02 Fineweb by pszemraj

By pszemraj · 12 downloads

Griffin C3t 8L V0.02 Fineweb is an open-source language model by pszemraj. Features: 1m LLM, VRAM: 0.7GB, License: apache-2.0, LLM Explorer Score: 0.17, Arc: 23.3, HellaSwag: 25.7, MMLU: 23.1.

Base model:finetune:pszemraj/g... Base model:pszemraj/griffin-v0... Dataset:bee-spoke-data/fineweb...   En   Endpoints compatible   Generated from trainer   Recurrent gemma   Region:us   Safetensors

Griffin C3t 8L V0.02 Fineweb Parameters and Internals

Model Type 
Pretraining experiment, recurrent_gemma architecture
Use Cases 
Limitations:
needs more training
Additional Notes 
The model seems to need more training to achieve better results.
Supported Languages 
en (proficient)
Training Details 
Data Sources:
BEE-spoke-data/fineweb-1M_en-med
Data Volume:
798621696 tokens
LLM NameGriffin C3t 8L V0.02 Fineweb
Repository πŸ€—https://huggingface.co/pszemraj/griffin-c3t-8L-v0.02-fineweb 
Base Model(s)  pszemraj/griffin-v0.01-c3t-8layer-simplewiki-silu   pszemraj/griffin-v0.01-c3t-8layer-simplewiki-silu
Model Size1m
Required VRAM0.7 GB
Updated2026-07-31
Maintainerpszemraj
Model Typerecurrent_gemma
Model Files  0.7 GB   0.0 GB
Supported Languagesen
Model ArchitectureRecurrentGemmaForCausalLM
Licenseapache-2.0
Model Max Length200000
Transformers Version4.40.1
Tokenizer ClassGPT2Tokenizer
Vocabulary Size65024
Torch Data Typefloat32

Best Alternatives to Griffin C3t 8L V0.02 Fineweb

Best Alternatives
Context / RAM
Downloads
Likes
...iffin Llama3t 8L V0.02 Fineweb0K / 0.9 GB110
Note: green Score (e.g. "73.2") means that the model is better than pszemraj/griffin-c3t-8L-v0.02-fineweb.