Training in progress, step 34500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0f8571f67cab7e7355d0a835675f10db403d911582b504d4460e834ec066221
+oid sha256:d4a6f0c25b4611f1c767905bf111be9724afb005e4306ef1bc73dc1c0e29cfa3
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e3e943eb7488886ee9103c8f80a97c673eb370cf20aa3e483dd2b7de7e0250c
+oid sha256:5226727ad5f179a9b000e5a412ba8dbddc5354e1baf3375db08313841c945818
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3eb6181ab2211380e8f9d27636b01f02a9625cef49eec6124c5f793a9f83748a
+oid sha256:8418f1a64c5a13294ff25d04acd41788e39a27614adf5076e870dc32b37a1b02
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:841b8722635c3f632de2d0434a474170c723a52d8927b56279dc71208dcd5fdc
+oid sha256:e8df234ef2da88e980e27d5e38430bdeab883c73aac61a65b34f9ea768e575f2
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f7263086319458c667c4ec9de24949320c079dc02cd6366f7b321d3562069b8
+oid sha256:57ac9006c5ceba86b32d9390a0bed65a2b226ca44ae334ed8b3b3374b8accf3e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94f277b3f91e9aeea511c74ffaa8a849317c65464dc1f9282b6b85fc9d3e68b8
+oid sha256:e6f7f395c95d2292d383b90683031f36e62fa6aed8aa58203f983b15524fc8f5
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 31000,
   "best_metric": 0.7226839661598206,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-31000",
-  "epoch": 2.615183447427121,
+  "epoch": 2.6536420275363435,
   "eval_steps": 250,
-  "global_step": 34000,
+  "global_step": 34500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5856,6 +5856,92 @@
       "eval_samples_per_second": 55.848,
       "eval_steps_per_second": 13.962,
       "step": 34000
+    },
+    {
+      "epoch": 2.619029305438043,
+      "grad_norm": 0.4978080093860626,
+      "learning_rate": 2.5790198166376645e-05,
+      "loss": 0.7309,
+      "step": 34050
+    },
+    {
+      "epoch": 2.6228751634489655,
+      "grad_norm": 0.779080331325531,
+      "learning_rate": 2.5530478144560168e-05,
+      "loss": 0.7105,
+      "step": 34100
+    },
+    {
+      "epoch": 2.626721021459888,
+      "grad_norm": 0.5153629779815674,
+      "learning_rate": 2.5275952523180012e-05,
+      "loss": 0.7377,
+      "step": 34150
+    },
+    {
+      "epoch": 2.63056687947081,
+      "grad_norm": 0.8356613516807556,
+      "learning_rate": 2.5016232501363528e-05,
+      "loss": 0.7135,
+      "step": 34200
+    },
+    {
+      "epoch": 2.6344127374817323,
+      "grad_norm": 0.5202348232269287,
+      "learning_rate": 2.475651247954705e-05,
+      "loss": 0.7333,
+      "step": 34250
+    },
+    {
+      "epoch": 2.6344127374817323,
+      "eval_loss": 0.7767261862754822,
+      "eval_runtime": 17.8651,
+      "eval_samples_per_second": 55.975,
+      "eval_steps_per_second": 13.994,
+      "step": 34250
+    },
+    {
+      "epoch": 2.6382585954926543,
+      "grad_norm": 0.9579488039016724,
+      "learning_rate": 2.4496792457730567e-05,
+      "loss": 0.7823,
+      "step": 34300
+    },
+    {
+      "epoch": 2.6421044535035767,
+      "grad_norm": 0.7704477906227112,
+      "learning_rate": 2.4237072435914084e-05,
+      "loss": 0.6979,
+      "step": 34350
+    },
+    {
+      "epoch": 2.6459503115144987,
+      "grad_norm": 0.8563690781593323,
+      "learning_rate": 2.3977352414097603e-05,
+      "loss": 0.7049,
+      "step": 34400
+    },
+    {
+      "epoch": 2.649796169525421,
+      "grad_norm": 0.663038432598114,
+      "learning_rate": 2.3717632392281123e-05,
+      "loss": 0.751,
+      "step": 34450
+    },
+    {
+      "epoch": 2.6536420275363435,
+      "grad_norm": 0.8598125576972961,
+      "learning_rate": 2.345791237046464e-05,
+      "loss": 0.6982,
+      "step": 34500
+    },
+    {
+      "epoch": 2.6536420275363435,
+      "eval_loss": 0.7752255201339722,
+      "eval_runtime": 17.9612,
+      "eval_samples_per_second": 55.675,
+      "eval_steps_per_second": 13.919,
+      "step": 34500
     }
   ],
   "logging_steps": 50,