Rename bf16_baseline to fp16_baseline (tensors are actually stored as GGML F16, not BF16)

Upload bf16_baseline/norm_stats.json with huggingface_hub

Upload bf16_baseline/tokenizer.model with huggingface_hub

Upload bf16_baseline/pi05.gguf with huggingface_hub

Upload awq_3bit/norm_stats.json with huggingface_hub

Upload awq_3bit/tokenizer.model with huggingface_hub

Upload awq_3bit/pi05.gguf with huggingface_hub

Delete awq_3bit/ur3_awq_v3l3_state.pt with huggingface_hub

Upload awq_3bit/manifest.json with huggingface_hub

Upload awq_3bit/ur3_awq_v3l3_state.pt with huggingface_hub

Upload actquant_3p0bpw/norm_stats.json with huggingface_hub

Upload actquant_3p0bpw/tokenizer.model with huggingface_hub

Upload actquant_3p0bpw/pi05.gguf with huggingface_hub

Upload actquant_2p5bpw/norm_stats.json with huggingface_hub

Upload actquant_2p5bpw/tokenizer.model with huggingface_hub

Upload actquant_2p5bpw/pi05.gguf with huggingface_hub

initial commit