-
Update username to leochen085; fix base checkpoint download snippet
-
Add process_sensor; fix remote-code imports
-
Cross-link the Llama-3.2-1B backbone variant (SLIP-Llama)
-
Update README.md
-
Update README.md
-
Create README.md
-
Pass dtype through to Gemma init for proper torch_dtype support
-
Init Gemma in float32, users control dtype via torch_dtype param
-
Remove Gemma dtype override
-
Re-save as uniform float32: caption.safetensors
-
Re-save as uniform float32: tsqa.safetensors
-
Re-save as uniform float32: ecg.safetensors
-
Re-save as uniform float32: sleep.safetensors
-
Re-save as uniform float32: har.safetensors
-
Re-save as uniform float32: model.safetensors
-
Uniform float32 weights, clean dtype handling: modeling_slip.py
-
Uniform float32 weights, clean dtype handling: config.json
-
Fix mixed-dtype handling in configuration_slip.py
-
Fix mixed-dtype handling in modeling_slip.py
-
Use set_default_dtype for uniform dtype init, assert batch size match in CrossAttention
-
Update configuration_slip.py: use HF torch_dtype instead of hardcoded bfloat16
-
Update modeling_slip.py: use HF torch_dtype instead of hardcoded bfloat16
-
Update config.json: use HF torch_dtype instead of hardcoded bfloat16
-
Fix CrossAttention batch size mismatch: expand context to match query batch
-
Fix dtype mismatches: cast entire model to bfloat16 in init
-
Remove old util/ directory
-
Remove old model_factory/ directory
-
Upload SLIP model, checkpoints, and source code
-
Upload SLIP model, checkpoints, and source code
-
initial commit
