alignment-handbook/zephyr-7b-sft-full
Full fine-tune vs mistralai/Mistral-7B-v0.1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 17/65, MLP 96/96, Attention 128/128, lm_head 1/1
13 derivatives checked
- MaziyarPanahi/zephyr-7b-sft-full-GGUF: Quantization
- jikaixuan/zephyr-ds: Adapter
- YuchenLi01/ultrafeedbackSkyworkAgree_alignmentZephyr7BSftFull_sdpo_score_ebs64_lr1e-07_0: Full fine-tune
- argilla/notus-7b-v1: Full fine-tune
- alignment-handbook/zephyr-7b-dpo-full: Full fine-tune
- UCLA-AGI/zephyr-7b-sft-full-SPIN-iter1: Full fine-tune
- NicholasCorrado/zephyr-7b-hh-dpo: Full fine-tune
- argilla/zephyr-7b-spin-iter0-v0: Full fine-tune
- weijie210/zephyr-7b-UC-0: Full fine-tune
- weijie210/zephyr-7b-UC-0-B0.5: Full fine-tune
- weijie210/zephyr-7b-dpo-maximal: Full fine-tune
- NicholasCorrado/tulu-2-7b-rlced-conifer-dpo: Full fine-tune
- TTTXXX01/zephyr-7b-DPOBS128-full: Full fine-tune
1 other repos hold byte-for-byte identical weight files.