mradermacher/ci-random-llama3-8b-GGUF

Same weights, converted vs hyper-accel/ci-random-llama3-8b. Same weights stored in another dtype (F16 → F16/F32), plus 1 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/5, MLP 0/6, Attention 0/8

On the Hub · Downloads on Model Pulse