tarekziade HF Staff
Publish MIC usage, slim fixture and reviewed Transformers expectations
d03ef8c verified | { | |
| "recorded_with": { | |
| "contract": "huggingface-mic/Jamba-tiny-dev-mic", | |
| "contract_name": "Jamba", | |
| "contract_revision": "7e840f9d924defee8bbe8fa47d0c69b50b714e0e", | |
| "framework": "transformers", | |
| "variant": "bfloat16", | |
| "target": "Jamba-tiny-dev", | |
| "kind": "checkpoint", | |
| "revision": "ed303361004ac875426a61675edecf8e9d976882", | |
| "environment": { | |
| "python": "3.11.18", | |
| "transformers": "5.22.0.dev0", | |
| "torch": "2.14.0+cu130", | |
| "torchvision": "0.29.2", | |
| "torchaudio": "2.11.0+cu130", | |
| "tokenizers": "0.23.3", | |
| "numpy": "2.4.6", | |
| "pillow": "13.0.0", | |
| "accelerate": "1.16.0.dev0", | |
| "safetensors": "0.8.0", | |
| "huggingface-hub": "2.3.0", | |
| "sentencepiece": "0.2.2", | |
| "protobuf": "7.36.2", | |
| "librosa": "0.11.0", | |
| "soundfile": "0.14.0", | |
| "scipy": "1.17.1" | |
| }, | |
| "seconds": 9.86, | |
| "environment_differs": { | |
| "torchvision": "0.29.2 (pinned 0.29.1)", | |
| "numpy": "2.4.6 (pinned 2.3.5)", | |
| "pillow": "13.0.0 (pinned 12.3.0)", | |
| "accelerate": "1.16.0.dev0 (pinned 1.15.0)" | |
| }, | |
| "platform": "cuda" | |
| }, | |
| "outputs": { | |
| "runtime": { | |
| "device": "cuda", | |
| "dtype": "bfloat16", | |
| "attn_implementation": "sdpa" | |
| }, | |
| "last_logits": [ | |
| [ | |
| -7.375, | |
| -7.375, | |
| 5.5625, | |
| -7.0, | |
| -7.3125 | |
| ], | |
| [ | |
| -7.78125, | |
| -7.9375, | |
| 4.84375, | |
| -7.40625, | |
| -7.75 | |
| ] | |
| ], | |
| "next_token_ids": [ | |
| 7754, | |
| 1808 | |
| ], | |
| "training": { | |
| "loss": 4.0856804847717285, | |
| "gradients": { | |
| "model.layers.0.mamba.in_proj.weight": true, | |
| "model.layers.0.feed_forward.gate_proj.weight": true | |
| }, | |
| "updated": true | |
| } | |
| } | |
| } | |