Checkpoint at step 80

checkpoints/{checkpoint-40 → checkpoint-60}/small_experts_and_gates.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcad30c967aa7b4708df7b4d5a68f0b6d8c6123a399b6c35b9e0a4e29e67abb8
+oid sha256:e934d0cf6af7ae2d9a784f24f2c7dd73c89d260d3b53c3d0e3d34ef537f65289
 size 206711911

checkpoints/{checkpoint-40 → checkpoint-80}/model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:114400ea3c1f6b4bf8985e9f050651f88d39f370453e65343bbaa9d1d34d65a2
+oid sha256:82affd842ec486fef2da08a6bc5620d759625dcd2a9671f7e1633db3e23e8154
 size 4998915096

checkpoints/{checkpoint-40 → checkpoint-80}/model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2231902fc66ad75a3b8f3686c1a68d275554007b4678fad5d73d74f364a78a16
+oid sha256:a982717a3e6e6f33e64b592ba98c998d77f1fea7b50229735c94b491857b47a7
 size 4998953920

checkpoints/{checkpoint-40 → checkpoint-80}/model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fec36ff27e7cffe29172cd7c3426cec7a63e832e4a36fd3dd8d4cf337082135e
+oid sha256:2321b2ac4c163adff676d015f90a88da265a1515998048140c100ab1b50a7643
 size 4046769808

checkpoints/{checkpoint-40 → checkpoint-80}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5592a107e3807fc0b70ae71b93a98f020ad60d21c3c068b7be1bbbac2001b89e
-size 391093394
+oid sha256:60afe906089703ce8ff6dc7db2e543bedd92d9c1554d99f956ac873fa29965c8
+size 394655075

checkpoints/{checkpoint-40 → checkpoint-80}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7b724dee474d2f28b5e7cd28f30ad6358e8e4bd90e6205eb9f6ebd8fb296e614
+oid sha256:d484e81fa22cdbcf66c2585a42950fa1db3e6db36dd8936a72f76c2844202008
 size 1465

checkpoints/{checkpoint-40 → checkpoint-80}/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.001962263225347535,
+  "epoch": 0.00392452645069507,
   "eval_steps": 500,
-  "global_step": 40,
+  "global_step": 80,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -36,6 +36,34 @@
       "learning_rate": 6.376716808371485e-07,
       "loss": 109.0438,
       "step": 40
+    },
+    {
+      "epoch": 0.002452829031684419,
+      "grad_norm": 53.75,
+      "learning_rate": 8.011772400261609e-07,
+      "loss": 106.0855,
+      "step": 50
+    },
+    {
+      "epoch": 0.0029433948380213027,
+      "grad_norm": 63.25,
+      "learning_rate": 9.646827992151733e-07,
+      "loss": 106.0407,
+      "step": 60
+    },
+    {
+      "epoch": 0.003433960644358187,
+      "grad_norm": 54.0,
+      "learning_rate": 1.1281883584041859e-06,
+      "loss": 104.9819,
+      "step": 70
+    },
+    {
+      "epoch": 0.00392452645069507,
+      "grad_norm": 51.25,
+      "learning_rate": 1.2916939175931983e-06,
+      "loss": 108.4838,
+      "step": 80
     }
   ],
   "logging_steps": 10,
@@ -55,7 +83,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.0882488276615168e+17,
+  "total_flos": 2.1764976553230336e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

logs/events.out.tfevents.1756831118.5624918391e3.1490.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a16f40df0bc3e351cb0924387161cfe2a0c9e6a01de958593980e5f0db6e1bb
-size 6486
+oid sha256:29a79c773a923a3f7506eff6045dcc232e3fb15f00e7e6cc002ad450cad87400
+size 6900