Checkpoint at step 60

checkpoints/{checkpoint-20 → checkpoint-40}/small_experts_and_gates.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0a1c589f5de5e24b6da0665fb82d1950a722bcadea7cbaab535b51df184e4bc
+oid sha256:dcad30c967aa7b4708df7b4d5a68f0b6d8c6123a399b6c35b9e0a4e29e67abb8
 size 206711911

checkpoints/{checkpoint-20 → checkpoint-60}/model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0da34e74ed72395fa85811978c57461923bcbdbb06f9470601de4746e227016a
+oid sha256:48b53e83b3b9e51270e2b63ef55313427723ff57c75e4cf4c8269e04de45b290
 size 4998915096

checkpoints/{checkpoint-20 → checkpoint-60}/model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98373ef59e208fd7f0484c3868fa3a90473d26efb42452c34146a9131d6ba93f
+oid sha256:7d944c12069808ce0f621767a0eec41762042e9c6daad49b55d161adf7b52702
 size 4998953920

checkpoints/{checkpoint-20 → checkpoint-60}/model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a51ed769a193f6f3e113a19b244a56b0f0d4dc4ae073aa3c7703199f6ecb3f89
+oid sha256:2b183dccf9c40fe45cbbaea5eed282b461eb701195ae659c5c2f3a6f73d10c88
 size 4046769808

checkpoints/{checkpoint-20 → checkpoint-60}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bafc6c5721f51d42e5675e5a4158ceb3aa937ac367c6910fdaf603bf11e56c1
-size 383574247
+oid sha256:f466719a427bface1fffa8fbb6d8e31641cd12d2939657c34d508b7693e82a30
+size 393467848

checkpoints/{checkpoint-20 → checkpoint-60}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:795c503600292c8c704d0cbf6e623255ed54848f02f567e3d52d4c939d5df78e
+oid sha256:6e1be529198179cd559ddcb4c59a9f665944a456be4a70f4f5dcf79350fe0534
 size 1465

checkpoints/{checkpoint-20 → checkpoint-60}/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0009811316126737675,
+  "epoch": 0.0029433948380213027,
   "eval_steps": 500,
-  "global_step": 20,
+  "global_step": 60,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -22,6 +22,34 @@
       "learning_rate": 3.106605624591236e-07,
       "loss": 106.5866,
       "step": 20
+    },
+    {
+      "epoch": 0.0014716974190106514,
+      "grad_norm": 52.75,
+      "learning_rate": 4.7416612164813603e-07,
+      "loss": 107.8943,
+      "step": 30
+    },
+    {
+      "epoch": 0.001962263225347535,
+      "grad_norm": 68.5,
+      "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
     }
   ],
   "logging_steps": 10,
@@ -41,7 +69,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.441244138307584e+16,
+  "total_flos": 1.6323732414922752e+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 sha256:6b893f207f0aecf3668043c1d7255468fc0a035f088b014a4f0616628302defc
-size 6072
+oid sha256:0a16f40df0bc3e351cb0924387161cfe2a0c9e6a01de958593980e5f0db6e1bb
+size 6486