LGAI-EXAONE/EXAONE-Deep-2.4B
Full fine-tune vs LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct. Fine-tune: 100% of the weights changed, across 30 of 30 layers.
Tensors changed: Attention 120/120, Norms 61/61, MLP 90/90, Embeddings 1/1
10 derivatives checked
- bartowski/LGAI-EXAONE_EXAONE-Deep-2.4B-GGUF: Same weights, converted
- mobilint/EXAONE-Deep-2.4B: Full fine-tune (fewer layers)
- second-state/EXAONE-Deep-2.4B-GGUF: Same weights, converted
- mobilint/EXAONE-Deep-2.4B-Batch16: Full fine-tune (fewer layers)
- mobilint/EXAONE-Deep-2.4B-regulus-rb-usb: Full fine-tune (fewer layers)
- LGAI-EXAONE/EXAONE-Deep-2.4B-GGUF: Same weights, converted
- gaianet/EXAONE-Deep-2.4B-GGUF: Same weights, converted
- DevQuasar-10/LGAI-EXAONE.EXAONE-Deep-2.4B-GGUF: Same weights, converted
- tensorblock/EXAONE-Deep-2.4B-GGUF: Quantization
- LGAI-EXAONE/EXAONE-Deep-2.4B-AWQ: Quantization
1 other repos hold byte-for-byte identical weight files.