meta-llama/Meta-Llama-3-70B-Instruct
Full fine-tune vs meta-llama/Meta-Llama-3-70B. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 161/161, MLP 240/240, Attention 320/320, lm_head 1/1
27 derivatives checked
- MaziyarPanahi/calme-2.2-llama3-70b: LoRA-style
- Nexusflow/Athene-70B: Full fine-tune
- second-state/Meta-Llama-3-70B-Instruct-GGUF: Same weights, converted
- QuantFactory/Meta-Llama-3-70B-Instruct-GGUF: Quantization
- LiteLLMs/Llama3-OpenBioLLM-70B-GGUF: Full fine-tune
- QuantFactory/Meta-Llama-3-70B-Instruct-GGUF-v2: Quantization
- qwp4w3hyb/Meta-Llama-3-70B-Instruct-iMat-GGUF: Same weights, converted
- bartowski/Meta-Llama-3-120B-Instruct-GGUF: Partial fine-tune (fewer layers)
- lmstudio-community/Meta-Llama-3-70B-Instruct-GGUF: Quantization
- DevQuasar-11/meta-llama.Meta-Llama-3-70B-Instruct-GGUF: Same weights, converted
- lmstudio-community/Meta-Llama-3-70B-Instruct-BPE-fix-GGUF: Quantization
- LiteLLMs/Meta-Llama-3-120B-Instruct-GGUF: Full fine-tune (fewer layers)
- TsinghuaC3I/Llama-3-70B-UltraMedical: Full fine-tune
- shenzhi-wang/Llama3-70B-Chinese-Chat-GGUF-4bit: Full fine-tune
- LoneStriker/OpenBioLLM-Llama3-70B-GGUF: Full fine-tune
- MaziyarPanahi/calme-2.4-llama3-70b: LoRA-style
- MaziyarPanahi/calme-2.3-llama3-70b: LoRA-style
- MaziyarPanahi/Llama-3-70B-Instruct-32k-v0.1: Identical copy
- shisa-ai/shisa-v1-llama3-70b: Full fine-tune
- DISLab/SummLlama3-70B: LoRA-style
- nisten/llamagnific-3-87b: Partial fine-tune
- llmixer/BigWeave-v31-96b: Partial fine-tune
- MaziyarPanahi/calme-2.1-llama3-70b: LoRA-style
- etemiz/Ostrich-70B-Llama3-251212: Full fine-tune
- some1nostr/Ostrich-Llama-3-70B: LoRA-style
- llmixer/BigWeave-v33-105b: Partial fine-tune (fewer layers)
- llmixer/BigWeave-v32-96b: Partial fine-tune
11 other repos hold byte-for-byte identical weight files.