nicholasKluge/Aira-2-774M

Partial fine-tune vs openai-community/gpt2-large. Partial change: 25% of the weights changed (attention, mlp, norms), in 9 of 36 layers.

Tensors changed: Attention 36/144, Norms 38/146, MLP 36/144, Embeddings 0/1

2 derivatives checked

• mradermacher/Aira-2-774M-i1-GGUF: Quantization
• mradermacher/Aira-2-774M-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse