AtlaAI/Selene-1-Mini-Llama-3.1-8B
Full fine-tune vs meta-llama/Llama-3.1-8B-Instruct. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 12/65, MLP 96/96, Attention 128/128, lm_head 1/1
15 derivatives checked
- bartowski/Selene-1-Mini-Llama-3.1-8B-GGUF: Same weights, converted
- mradermacher/Selene-1-Mini-Llama-3.1-8B-i1-GGUF: Quantization
- AtlaAI/Selene-1-Mini-Llama-3.1-8B-Q4_K_M-GGUF: Quantization
- mradermacher/Selene-1-Mini-Llama-3.1-8B-GGUF: Same weights, converted
- tensorblock/Selene-1-Mini-Llama-3.1-8B-GGUF: Quantization
- DevQuasar-4/AtlaAI.Selene-1-Mini-Llama-3.1-8B-GGUF: Same weights, converted
- mergekit-community/L3.1-Athena-a-8B: Full fine-tune
- mergekit-community/L3.1-Athena-j-8B: Full fine-tune
- mergekit-community/L3.1-Athena-k-8B: Full fine-tune
- mergekit-community/L3.1-Athena-g-8B: Full fine-tune
- mergekit-community/L3.1-Athena-h-8B: Full fine-tune
- mergekit-community/L3.1-Athena-l-8B: Full fine-tune
- mergekit-community/L3.1-Athena-f-8B: Full fine-tune
- mergekit-community/L3.1-Athena-l2-8B: Full fine-tune
- mergekit-community/L3.1-Athena-l3-8B: Full fine-tune
1 other repos hold byte-for-byte identical weight files.