Disya/Qwen3-8b-Base-RP-slerp

Full fine-tune vs allura-org/remnant-qwen3-8b. Fine-tune: 90% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 0/1, Norms 117/145, MLP 105/108, Attention 140/144, lm_head 1/1

2 derivatives checked

• mradermacher/Qwen3-8b-Base-RP-slerp-i1-GGUF: Quantization
• mradermacher/Qwen3-8b-Base-RP-slerp-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse