dphn/dolphin-2.9-llama3-8b
Full fine-tune vs meta-llama/Meta-Llama-3-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
21 derivatives checked
- bartowski/dolphin-2.9-llama3-8b-GGUF: Same weights, converted
- stephenlzc/dolphin-llama3-zh-cn-uncensored: LoRA-style
- QuantFactory/dolphin-2.9-llama3-8b-GGUF: Quantization
- foxtail2/dolphin-2.9-llama3-8b-GGUF: Same weights, converted
- mradermacher/dolphin-2.9-llama3-8b-i1-GGUF: Quantization
- mradermacher/dolphin-2.9-llama3-8b-GGUF: Same weights, converted
- m2hgamerz/m2hai-unsensored-8b: Quantization
- cstr/dolphin-2.9-llama3-8b-GGUF: Quantization
- markhneedham/dolphin-2.9-llama3-8b-Q4_K_M-GGUF: Same weights, converted
- Cutyp/dolphin-8b-merged: Same weights, repacked
- kafirind17/dolphin-2.9-llama3-8b-GGUF: Same weights, converted
- bigsexbigflex/pineapple: Full fine-tune
- seyf1elislam/dolphin-2.9-llama3-8b-GGUF: Quantization
- tensorblock/dolphin-2.9-llama3-8b-GGUF: Quantization
- khjvgvyfc/dolphin-2.9-llama3-8b-GGUF: Quantization
- itlwas/dolphin-2.9-llama3-8b-Q4_K_M-GGUF: Quantization
- nbeerbower/llama-3-slerp-dolphin-sauce-8B: Full fine-tune
- timpal0l/dolphin-2.9-llama3-8b-flashback: Full fine-tune
- ChaoticNeutrals/Poppy_Porpoise-v0.4-L3-8B: Full fine-tune
- bibidentuhanoi/BMO-7B-Instruct_3.1: LoRA-style
- mergekit-community/mergekit-slerp-zwbosgo: Full fine-tune
11 other repos hold byte-for-byte identical weight files.