mpasila/Mistral-7B-Holodeck-1-Instruct-32k

Full fine-tune vs KoboldAI/Mistral-7B-Holodeck-1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128, lm_head 1/1

2 derivatives checked

• mradermacher/Mistral-7B-Holodeck-1-Instruct-32k-i1-GGUF: Quantization
• mradermacher/Mistral-7B-Holodeck-1-Instruct-32k-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse