Training in progress, step 30000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38b5f2ea569041576fa8587161c76c0f3c5a7dd581ad16d3677954bc1f4d9e6e
+oid sha256:b53972236677d35c7627b344577268d19eab8e86becdb001ad3d20c33562409f
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:168d5bd2debf9851704b7f4b3a599e4063b99ac6ab8d0b5d3299ef76cdfaee3b
+oid sha256:9ddb71187cf05b2d152abcaf17e6a1a66bbaf60093ba03650f5df78e882a41ae
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97942529c6161fa75e24d4bfa6d4f321f66089ea579319270d381c757a3814a9
+oid sha256:0caa256a01d13ade6a76529796c9eea8cb685776ddc6a24c1e4a28d00ce8ca63
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ec2e05a13a3420f603b9723227d4fcdae2660d4fc99eab774732776aece45a9
-size 14244
+oid sha256:fd6900b77a3aaa53dbd9ff02530a8bdbceb59bb2b9ea31ba74caa08f3dcac033
+size 14308

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1849a7944c4c7ac6e3a5cbc512aa60f489a5f5e80169ce90a617a00d1a8a86e
+oid sha256:e8c04c082df1347d4a5bd7568aa1ea9e76f0f21cd9fd31e93e6a215b3ffd653d
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8946eb030e4e86293b4ece2904285416cc1c901bf3058e686e13ff8f169c0a22
+oid sha256:7472cb4cae0bcbbe91407a8ab253d568120f4cb09f676c536e7c370fc9153e84
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 29250,
-  "best_metric": 0.7279431223869324,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-29000",
-  "epoch": 2.26905622644412,
+  "best_global_step": 30000,
+  "best_metric": 0.7264938354492188,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-30000",
+  "epoch": 2.307514806553342,
   "eval_steps": 250,
-  "global_step": 29500,
+  "global_step": 30000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5082,6 +5082,92 @@
       "eval_samples_per_second": 55.949,
       "eval_steps_per_second": 13.987,
       "step": 29500
+    },
+    {
+      "epoch": 2.272902084455042,
+      "grad_norm": 0.7172055840492249,
+      "learning_rate": 4.916500012986001e-05,
+      "loss": 0.6984,
+      "step": 29550
+    },
+    {
+      "epoch": 2.2767479424659642,
+      "grad_norm": 0.6817266345024109,
+      "learning_rate": 4.890528010804353e-05,
+      "loss": 0.7059,
+      "step": 29600
+    },
+    {
+      "epoch": 2.280593800476886,
+      "grad_norm": 1.0110056400299072,
+      "learning_rate": 4.864556008622705e-05,
+      "loss": 0.7304,
+      "step": 29650
+    },
+    {
+      "epoch": 2.2844396584878086,
+      "grad_norm": 0.6823923587799072,
+      "learning_rate": 4.838584006441057e-05,
+      "loss": 0.7462,
+      "step": 29700
+    },
+    {
+      "epoch": 2.288285516498731,
+      "grad_norm": 0.7316113710403442,
+      "learning_rate": 4.812612004259408e-05,
+      "loss": 0.7068,
+      "step": 29750
+    },
+    {
+      "epoch": 2.288285516498731,
+      "eval_loss": 0.7273894548416138,
+      "eval_runtime": 17.8859,
+      "eval_samples_per_second": 55.91,
+      "eval_steps_per_second": 13.978,
+      "step": 29750
+    },
+    {
+      "epoch": 2.292131374509653,
+      "grad_norm": 1.200492024421692,
+      "learning_rate": 4.78664000207776e-05,
+      "loss": 0.6887,
+      "step": 29800
+    },
+    {
+      "epoch": 2.2959772325205754,
+      "grad_norm": 1.2268471717834473,
+      "learning_rate": 4.760667999896112e-05,
+      "loss": 0.7259,
+      "step": 29850
+    },
+    {
+      "epoch": 2.299823090531498,
+      "grad_norm": 0.7251473069190979,
+      "learning_rate": 4.734695997714464e-05,
+      "loss": 0.6931,
+      "step": 29900
+    },
+    {
+      "epoch": 2.3036689485424198,
+      "grad_norm": 0.5327921509742737,
+      "learning_rate": 4.708723995532816e-05,
+      "loss": 0.7447,
+      "step": 29950
+    },
+    {
+      "epoch": 2.307514806553342,
+      "grad_norm": 0.7111462950706482,
+      "learning_rate": 4.6827519933511674e-05,
+      "loss": 0.7221,
+      "step": 30000
+    },
+    {
+      "epoch": 2.307514806553342,
+      "eval_loss": 0.7264938354492188,
+      "eval_runtime": 17.7847,
+      "eval_samples_per_second": 56.228,
+      "eval_steps_per_second": 14.057,
+      "step": 30000
     }
   ],
   "logging_steps": 50,