Upload training_logs/it_tfa_big__seed42.json with huggingface_hub

training_logs/it_tfa_big__seed42.json

@@ -0,0 +1,125 @@
+{
+  "loss": [
+    2721.08935546875,
+    1053.83203125,
+    1007.916259765625,
+    996.9671020507812,
+    979.9942626953125,
+    965.4766845703125,
+    951.537353515625,
+    953.679931640625,
+    932.845703125,
+    943.7442016601562,
+    931.8909912109375,
+    914.9315185546875,
+    902.3789672851562,
+    907.9058227539062,
+    891.9739990234375,
+    887.3131103515625,
+    876.37939453125,
+    865.1234130859375,
+    873.8612060546875,
+    883.999267578125,
+    874.157470703125,
+    844.5355224609375,
+    839.478515625,
+    862.7964477539062,
+    856.760498046875,
+    863.526123046875,
+    835.718017578125
+  ],
+  "l0": [
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0,
+    500.0
+  ],
+  "steps_logged": [
+    0,
+    200,
+    400,
+    600,
+    800,
+    1000,
+    1200,
+    1400,
+    1600,
+    1800,
+    2000,
+    2200,
+    2400,
+    2600,
+    2800,
+    3000,
+    3200,
+    3400,
+    3600,
+    3800,
+    4000,
+    4200,
+    4400,
+    4600,
+    4800,
+    5000,
+    5200
+  ],
+  "final_step": 5200,
+  "converged": true,
+  "plateau_last": 0.01920854087932094,
+  "elapsed_s": 7397.639723777771,
+  "scaling_factor": 0.22266921934949058,
+  "skipped_steps": 0,
+  "row": 7,
+  "arch_id": "tfa_big",
+  "arch": "tfa_big",
+  "group": 1,
+  "src_class": "TemporalSAE",
+  "src_module": "src.architectures._tfa_module",
+  "T": null,
+  "T_max": null,
+  "t_sample": null,
+  "n_layers": null,
+  "k_win": 500,
+  "k_pos": null,
+  "shifts": null,
+  "alpha": null,
+  "gamma": null,
+  "n_scales": null,
+  "seed": 42,
+  "d_in": 2304,
+  "d_sae": 18432,
+  "subject_model": "google/gemma-2-2b-it",
+  "anchor_layer": 13,
+  "mlc_layers": [
+    11,
+    12,
+    13,
+    14,
+    15
+  ],
+  "phase": "phase7_unification",
+  "run_id": "it_tfa_big__seed42"
+}