trollek/ThoughtStream-4B-v0.3
LoRA-style vs h2oai/h2o-danube3-4b-base. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 24 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/49, MLP 72/72, Attention 96/96
2 derivatives checked
- mradermacher/ThoughtStream-4B-v0.3-GGUF: Same weights, converted
- trollek/ThoughtStream-4B-v0.3-GGUF: Same weights, converted