Checkpoint at step 100

checkpoints/{checkpoint-60 → checkpoint-100}/model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48b53e83b3b9e51270e2b63ef55313427723ff57c75e4cf4c8269e04de45b290
+oid sha256:306ae2c5a62a9d5985833823942be31e4af511252576cd26b40a35b7d9201283
 size 4998915096

checkpoints/{checkpoint-60 → checkpoint-100}/model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d944c12069808ce0f621767a0eec41762042e9c6daad49b55d161adf7b52702
+oid sha256:3799244485da1d90d05ca0b89bec8a7fe2e5476a63c730756f9b69c4e67d4e6f
 size 4998953920

checkpoints/{checkpoint-60 → checkpoint-100}/model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b183dccf9c40fe45cbbaea5eed282b461eb701195ae659c5c2f3a6f73d10c88
+oid sha256:4bb7bd262681ea7437fa4a014046e12ecb56c73fd62debb8d2d232f416305560
 size 4046769808

checkpoints/{checkpoint-60 → checkpoint-100}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f466719a427bface1fffa8fbb6d8e31641cd12d2939657c34d508b7693e82a30
-size 393467848
+oid sha256:a81c59ba9675a87d095d8a3b1b6affbaa56688ab72a82724c47d00c8b7f93657
+size 395050796

checkpoints/{checkpoint-60 → checkpoint-100}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e1be529198179cd559ddcb4c59a9f665944a456be4a70f4f5dcf79350fe0534
+oid sha256:528392556c02087909bc6166c540e006acf7c36a66f3053e5688b159ec58a0e0
 size 1465

checkpoints/{checkpoint-60 → checkpoint-100}/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0029433948380213027,
+  "epoch": 0.004905658063368838,
   "eval_steps": 500,
-  "global_step": 60,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -50,6 +50,34 @@
       "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
+    },
+    {
+      "epoch": 0.0044150922570319545,
+      "grad_norm": 59.75,
+      "learning_rate": 1.4551994767822106e-06,
+      "loss": 104.6467,
+      "step": 90
+    },
+    {
+      "epoch": 0.004905658063368838,
+      "grad_norm": 53.25,
+      "learning_rate": 1.618705035971223e-06,
+      "loss": 105.9295,
+      "step": 100
     }
   ],
   "logging_steps": 10,
@@ -69,7 +97,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.6323732414922752e+17,
+  "total_flos": 2.720622069153792e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

checkpoints/{checkpoint-60 → checkpoint-80}/small_experts_and_gates.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e934d0cf6af7ae2d9a784f24f2c7dd73c89d260d3b53c3d0e3d34ef537f65289
+oid sha256:d18f5bc94ce657ff2622e730419e52a30ceebcb5e01cdd278338e0e14c0f1077
 size 206711911

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29a79c773a923a3f7506eff6045dcc232e3fb15f00e7e6cc002ad450cad87400
-size 6900
+oid sha256:10f992425880062aee94cbc0044c2e5c836f3e78a7ddc54153790b6921b487bb
+size 7314