Training in progress, step 29500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f7e588a197a8570159058327c70025893ca766d0fbd2a25bb0403e673011256
+oid sha256:38b5f2ea569041576fa8587161c76c0f3c5a7dd581ad16d3677954bc1f4d9e6e
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84eb396661bc75d364b2d44d83a4c4960855665d430d64c5097382769b27773b
+oid sha256:168d5bd2debf9851704b7f4b3a599e4063b99ac6ab8d0b5d3299ef76cdfaee3b
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95894271fa4fbddfeacbe1ad5d29056d6607d4cbede968d833f230635c10694c
+oid sha256:97942529c6161fa75e24d4bfa6d4f321f66089ea579319270d381c757a3814a9
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bdc223e0883d03981b825b4fc5e59f6538251017a05a433c157f563714a2f9ba
-size 14180
+oid sha256:0ec2e05a13a3420f603b9723227d4fcdae2660d4fc99eab774732776aece45a9
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24e0d66d723576c324a7fe34e5dbff8e5703e0dde1366b79228a8fb754fe8fd9
+oid sha256:b1849a7944c4c7ac6e3a5cbc512aa60f489a5f5e80169ce90a617a00d1a8a86e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9c653c5f990cfe923931e3bba4aa0b8bc89aeef6f3280f590db66eda67a330f
+oid sha256:8946eb030e4e86293b4ece2904285416cc1c901bf3058e686e13ff8f169c0a22
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 29000,
-  "best_metric": 0.7284311652183533,
+  "best_global_step": 29250,
+  "best_metric": 0.7279431223869324,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-29000",
-  "epoch": 2.230597646334897,
+  "epoch": 2.26905622644412,
   "eval_steps": 250,
-  "global_step": 29000,
+  "global_step": 29500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4996,6 +4996,92 @@
       "eval_samples_per_second": 55.882,
       "eval_steps_per_second": 13.97,
       "step": 29000
+    },
+    {
+      "epoch": 2.2344435043458195,
+      "grad_norm": 0.45148393511772156,
+      "learning_rate": 5.1762200348024836e-05,
+      "loss": 0.7068,
+      "step": 29050
+    },
+    {
+      "epoch": 2.238289362356742,
+      "grad_norm": 0.4047794044017792,
+      "learning_rate": 5.150248032620835e-05,
+      "loss": 0.7015,
+      "step": 29100
+    },
+    {
+      "epoch": 2.242135220367664,
+      "grad_norm": 0.8585315942764282,
+      "learning_rate": 5.124276030439187e-05,
+      "loss": 0.7028,
+      "step": 29150
+    },
+    {
+      "epoch": 2.2459810783785863,
+      "grad_norm": 0.5665230751037598,
+      "learning_rate": 5.098304028257539e-05,
+      "loss": 0.658,
+      "step": 29200
+    },
+    {
+      "epoch": 2.2498269363895087,
+      "grad_norm": 0.7042600512504578,
+      "learning_rate": 5.07233202607589e-05,
+      "loss": 0.6938,
+      "step": 29250
+    },
+    {
+      "epoch": 2.2498269363895087,
+      "eval_loss": 0.7279431223869324,
+      "eval_runtime": 17.8253,
+      "eval_samples_per_second": 56.1,
+      "eval_steps_per_second": 14.025,
+      "step": 29250
+    },
+    {
+      "epoch": 2.2536727944004307,
+      "grad_norm": 0.7422949075698853,
+      "learning_rate": 5.046360023894242e-05,
+      "loss": 0.6996,
+      "step": 29300
+    },
+    {
+      "epoch": 2.257518652411353,
+      "grad_norm": 1.0139210224151611,
+      "learning_rate": 5.020388021712594e-05,
+      "loss": 0.7311,
+      "step": 29350
+    },
+    {
+      "epoch": 2.261364510422275,
+      "grad_norm": 0.5937057137489319,
+      "learning_rate": 4.994416019530946e-05,
+      "loss": 0.704,
+      "step": 29400
+    },
+    {
+      "epoch": 2.2652103684331975,
+      "grad_norm": 1.064329981803894,
+      "learning_rate": 4.968444017349297e-05,
+      "loss": 0.6792,
+      "step": 29450
+    },
+    {
+      "epoch": 2.26905622644412,
+      "grad_norm": 0.9638292789459229,
+      "learning_rate": 4.94247201516765e-05,
+      "loss": 0.74,
+      "step": 29500
+    },
+    {
+      "epoch": 2.26905622644412,
+      "eval_loss": 0.7287396192550659,
+      "eval_runtime": 17.8735,
+      "eval_samples_per_second": 55.949,
+      "eval_steps_per_second": 13.987,
+      "step": 29500
     }
   ],
   "logging_steps": 50,