Michael Yang adff143bcd fix: mllama quality (#10807) ...

* fix mllama convert

- transform attn_gate and ffn_gate
- swap attention heads for vision models

* fix mllama

the mlp gate which was applied in the wrong place

2025-05-22 11:30:49 -07:00

..

sentencepiece

…

testdata

…

convert_bert.go

…

convert_commandr.go

…

convert_gemma2_adapter.go

…

convert_gemma2.go

…

convert_gemma3.go

…

convert_gemma.go

…

convert_llama4.go

…

convert_llama_adapter.go

…

convert_llama.go

fix mllama conversion (#10716)

2025-05-15 12:15:01 -07:00

convert_mistral.go

…

convert_mixtral.go

…

convert_mllama.go

fix: mllama quality (#10807)

2025-05-22 11:30:49 -07:00

convert_phi3.go

…

convert_qwen2.go

…

convert_qwen25vl.go

…

convert_test.go

ggml: Seperate tensor load from backend creation

2025-05-19 09:54:22 -07:00

convert.go

model: handle multiple eos tokens (#10577)

2025-05-16 13:40:23 -07:00

reader_safetensors.go

…

reader_torch.go

…

reader.go

…

sentencepiece_model.proto

…

tensor.go

…

tokenizer_spm.go

…

tokenizer_test.go

model: handle multiple eos tokens (#10577)

2025-05-16 13:40:23 -07:00

tokenizer.go

model: handle multiple eos tokens (#10577)

2025-05-16 13:40:23 -07:00