prithivMLmods/Megatron-Corpus-14B-Exp
LoRA-style vs prithivMLmods/Megatron-Opus-14B-Exp. 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 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 144/144, Attention 192/336, lm_head 0/1
5 derivatives checked
- mradermacher/Megatron-Corpus-14B-Exp-i1-GGUF: Quantization
- mradermacher/Megatron-Corpus-14B-Exp-GGUF: Quantization
- prithivMLmods/Elita-1: Full fine-tune
- prithivMLmods/Magellanic-Opus-14B-Exp: Full fine-tune
- prithivMLmods/Megatron-Corpus-14B-Exp.v2: Full fine-tune