Instructions to use vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download sequential_progress.json from vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion: direct link, hf CLI and curl.
- Browser
- Download file 2.54 kB
-
https://huggingface.co/vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion/resolve/main/sequential_progress.json
- Command line
-
hf download hf://vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion/sequential_progress.json
-
curl -L -o sequential_progress.json https://huggingface.co/vtava/functiongemma-270m-it-simple-tool-calling-memory-fusion/resolve/main/sequential_progress.json
2.54 kB
| { | |
| "format_version": 1, | |
| "stage": "accepted_layer_17", | |
| "target_layers": [ | |
| 5, | |
| 11, | |
| 17 | |
| ], | |
| "accepted_layers": [ | |
| 5, | |
| 11, | |
| 17 | |
| ], | |
| "config": { | |
| "feature_dim": 32, | |
| "memory_rank": 64, | |
| "dilations": [ | |
| 1, | |
| 2, | |
| 4, | |
| 8, | |
| 16, | |
| 32, | |
| 64, | |
| 128 | |
| ], | |
| "shifted_window": 8, | |
| "train_output_projection": true | |
| }, | |
| "layer_reports": [ | |
| { | |
| "layer": 5, | |
| "accepted": false, | |
| "steps": 50, | |
| "nmse": 0.39167162775993347, | |
| "cosine": 0.77734375, | |
| "probe_nll": 6.578028440475464, | |
| "incremental_delta_nll": -1.947494626045227, | |
| "cumulative_delta_nll": -1.947494626045227, | |
| "round": 1 | |
| }, | |
| { | |
| "layer": 5, | |
| "accepted": false, | |
| "steps": 175, | |
| "nmse": 0.24159154295921326, | |
| "cosine": 0.8691575527191162, | |
| "probe_nll": 6.600935220718384, | |
| "incremental_delta_nll": -1.9245878458023071, | |
| "cumulative_delta_nll": -1.9245878458023071, | |
| "round": 2 | |
| }, | |
| { | |
| "layer": 5, | |
| "accepted": true, | |
| "steps": 225, | |
| "nmse": 0.15754102170467377, | |
| "cosine": 0.922616183757782, | |
| "probe_nll": 6.462509989738464, | |
| "incremental_delta_nll": -2.0630130767822266, | |
| "cumulative_delta_nll": -2.0630130767822266, | |
| "round": 3 | |
| }, | |
| { | |
| "layer": 11, | |
| "accepted": true, | |
| "steps": 150, | |
| "nmse": 0.16000066697597504, | |
| "cosine": 0.9214716553688049, | |
| "probe_nll": 6.223196983337402, | |
| "incremental_delta_nll": -0.239313006401062, | |
| "cumulative_delta_nll": -2.3023260831832886, | |
| "round": 1 | |
| }, | |
| { | |
| "layer": 17, | |
| "accepted": false, | |
| "steps": 300, | |
| "nmse": 0.15326528251171112, | |
| "cosine": 0.9228067398071289, | |
| "probe_nll": 6.2936941385269165, | |
| "incremental_delta_nll": 0.07049715518951416, | |
| "cumulative_delta_nll": -2.2318289279937744, | |
| "round": 1 | |
| }, | |
| { | |
| "layer": 17, | |
| "accepted": false, | |
| "steps": 225, | |
| "nmse": 0.11938813328742981, | |
| "cosine": 0.9484158754348755, | |
| "probe_nll": 6.2485164403915405, | |
| "incremental_delta_nll": 0.025319457054138184, | |
| "cumulative_delta_nll": -2.2770066261291504, | |
| "round": 2 | |
| }, | |
| { | |
| "layer": 17, | |
| "accepted": true, | |
| "steps": 225, | |
| "nmse": 0.13086512684822083, | |
| "cosine": 0.9407033920288086, | |
| "probe_nll": 6.230323672294617, | |
| "incremental_delta_nll": 0.0071266889572143555, | |
| "cumulative_delta_nll": -2.295199394226074, | |
| "round": 3 | |
| } | |
| ] | |
| } |