Alberto-Codes/Llama-3_3-Nemotron-Super-49B-v1_5-sensitivity-maps
Llama-3_3-Nemotron-Super-49B-v1_5-sensitivity-maps This dataset carries the per-layer quantization sensitivity maps of nvidia/Llama-3_3-Nemotron-Super-49B-v1_5. vramfit measured them. A sensitivity map records one damage number per layer group and candidate precision. Damage is the shift in the model's output distribution when that group alone quantizes — mean final-logits KL divergence against the bf16 reference. The maps describe the base model, not any quantized file. They… See the full description on the dataset page: https://huggingface.co/datasets/Alberto-Codes/Llama-3_3-Nemotron-Super-49B-v1_5-sensitivity-maps.
Card: state the shipped headroom in the plan command (#489)
Revise the card for the key rename (#121)
Rename the envelope keys to vramfit (#118, #121)
Upload README.md with huggingface_hub
Upload README.md with huggingface_hub
Upload calibration.txt with huggingface_hub
Upload sensitivity-64k-kquant-imx-no2-sized.json with huggingface_hub
Upload sensitivity-64k-kquant-imx-no2.json with huggingface_hub
Upload sensitivity-64k-kquant-imx-sized.json with huggingface_hub
Upload sensitivity-64k-kquant-imx.runlog.jsonl with huggingface_hub
Upload sensitivity-64k-kquant-imx.json with huggingface_hub
Upload sensitivity-64k-kquant.runlog.jsonl with huggingface_hub
Upload sensitivity-64k-kquant.json with huggingface_hub
Upload sensitivity-64k.runlog.jsonl with huggingface_hub
Upload sensitivity-64k.json with huggingface_hub
Upload sensitivity-32k.runlog.jsonl with huggingface_hub
Upload sensitivity-32k.json with huggingface_hub
Upload sensitivity-8k.runlog.jsonl with huggingface_hub
Upload sensitivity-8k.json with huggingface_hub
initial commit