Unsloth 4-bit Dynamic Quants Collection Unsloths Dynamic 4bit Quants selectively skips quantizing certain parameters; greatly improving accuracy while only using <10% more VRAM than BnB 4bit β’ 18 items β’ Updated 26 minutes ago β’ 37
DolphinLabeled Datasets Collection Eric Hartford has added labels to help you filter datasets, for your pleasure. β’ 5 items β’ Updated about 1 month ago β’ 11
DPO Collection Various useful datasets with preference optimization β’ 16 items β’ Updated 13 days ago β’ 3
FuseChat 3.0 Collection Preference Optimization for Implicit Model Fusion β’ 11 items β’ Updated Dec 12, 2024 β’ 10
EXAONE-3.5 Collection EXAONE 3.5 language model series including instruction-tuned models of 2.4B, 7.8B, and 32B. β’ 10 items β’ Updated Dec 10, 2024 β’ 90
view article Article πΊπ¦ββ¬ LLM Comparison/Test: 25 SOTA LLMs (including QwQ) through 59 MMLU-Pro CS benchmark runs By wolfram β’ Dec 4, 2024 β’ 76
Celeste Collection Series of roleplay specialist models with vibrant, diverse and human-like prose β’ 6 items β’ Updated Aug 8, 2024 β’ 7
Gemma 2 2B Release Collection The 2.6B parameter version of Gemma 2. β’ 6 items β’ Updated Dec 13, 2024 β’ 78