Add capability-head metadata
Browse files- capability_head_config.json +29 -0
capability_head_config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"format_version": 1,
|
| 3 |
+
"head_type": "capability",
|
| 4 |
+
"base_model": "google/gemma-4-E4B-it",
|
| 5 |
+
"model_family": "gemma4",
|
| 6 |
+
"thinking_mode": "off",
|
| 7 |
+
"variant": "full-trajectory",
|
| 8 |
+
"architecture": {
|
| 9 |
+
"num_labels": 1,
|
| 10 |
+
"head_input_mode": "completion_text_only",
|
| 11 |
+
"hidden_encoder_type": "lite",
|
| 12 |
+
"selected_hidden_layer_indices": null,
|
| 13 |
+
"hidden_layer_selection": "last",
|
| 14 |
+
"hidden_layer_index": null,
|
| 15 |
+
"hidden_layer_indices": null,
|
| 16 |
+
"max_seq_len": 32000,
|
| 17 |
+
"max_pixels": 200000,
|
| 18 |
+
"dtype": "bf16",
|
| 19 |
+
"aux_label_column": "correctness_score",
|
| 20 |
+
"regression_loss": "mse",
|
| 21 |
+
"failure_threshold": 0.5
|
| 22 |
+
},
|
| 23 |
+
"weight_file": "capability_head.pt",
|
| 24 |
+
"weight_sha256": "624f4cf4a8df7ca686fbc710dad6262f43c910ed46a7a1a931b7a7357d1fc3aa",
|
| 25 |
+
"parameter_count": 2863517,
|
| 26 |
+
"paper_use": "Table 2 routing and Table 4 web-search control",
|
| 27 |
+
"code": "https://github.com/Amirhosein-gh98/Multi-Head-Latent-Control",
|
| 28 |
+
"paper": "https://arxiv.org/abs/2607.14277"
|
| 29 |
+
}
|