Training in progress, step 17000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d2bb37f0bb2536ca33711abe7b3143ef3974ef2ca0c218977672374d7ca8e27
+oid sha256:b80b3583e2afb611f4c9adf5dda97e41fa3abaeec3239ac74a43ddde70a05578
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3916cf9823a85ba87d1391c608e5091a2091feb324e91b9bb2a76ddf86c0982d
+oid sha256:dfba693cf1d282bbea5c304bdd309307768a9ce459d448905954c71c856485be
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:518b6ef8fc143d463491edf1260763c10b5e11e873cd36abf76a046091e577c9
+oid sha256:ba8347dc980899d1fbf4c0d68fa8a8cf6d3395e6ddd6aee4b5d0bea2791716a0
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d28d5f3a49ada942ec355f276b33d6e4bff345075b872b6e2b651f5666a06f0a
+oid sha256:0a9d5359cc143881ecdb88768e105d086fd2336ed10ecf8c85dd03dc0e505da1
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b340848445b4f5b2d5567ddb4c818db5bf89137348148d19bd0b15ff0d69c43d
+oid sha256:d3333c3c82f9e4ec8ba5829fe0ea7dc977fdffe09f41702806e9515db920536e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d72187e2b5b8f8cffa32a0fbea90f82e23c165518eb95af21f7e18bd9e8c4a42
+oid sha256:14c33e4d9235be55afaa758d537d857af11e671e33d9253fad9c2c79b3373700
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 16250,
-  "best_metric": 1.4893407821655273,
+  "best_global_step": 16750,
+  "best_metric": 1.4798808097839355,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-15500",
-  "epoch": 1.2691331436043383,
+  "epoch": 1.3075917237135606,
   "eval_steps": 250,
-  "global_step": 16500,
+  "global_step": 17000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2846,6 +2846,92 @@
       "eval_samples_per_second": 55.967,
       "eval_steps_per_second": 13.992,
       "step": 16500
+    },
+    {
+      "epoch": 1.2729790016152602,
+      "grad_norm": 1.3523976802825928,
+      "learning_rate": 0.00011668181700127263,
+      "loss": 1.4796,
+      "step": 16550
+    },
+    {
+      "epoch": 1.2768248596261826,
+      "grad_norm": 1.9770869016647339,
+      "learning_rate": 0.00011642209697945614,
+      "loss": 1.4756,
+      "step": 16600
+    },
+    {
+      "epoch": 1.2806707176371048,
+      "grad_norm": 1.7973159551620483,
+      "learning_rate": 0.00011616237695763967,
+      "loss": 1.4163,
+      "step": 16650
+    },
+    {
+      "epoch": 1.284516575648027,
+      "grad_norm": 1.3054739236831665,
+      "learning_rate": 0.00011590265693582318,
+      "loss": 1.4105,
+      "step": 16700
+    },
+    {
+      "epoch": 1.2883624336589494,
+      "grad_norm": 1.457047939300537,
+      "learning_rate": 0.00011564293691400672,
+      "loss": 1.4698,
+      "step": 16750
+    },
+    {
+      "epoch": 1.2883624336589494,
+      "eval_loss": 1.4798808097839355,
+      "eval_runtime": 17.8696,
+      "eval_samples_per_second": 55.961,
+      "eval_steps_per_second": 13.99,
+      "step": 16750
+    },
+    {
+      "epoch": 1.2922082916698716,
+      "grad_norm": 2.117663860321045,
+      "learning_rate": 0.00011538841129262656,
+      "loss": 1.4252,
+      "step": 16800
+    },
+    {
+      "epoch": 1.2960541496807938,
+      "grad_norm": 3.7321341037750244,
+      "learning_rate": 0.00011512869127081007,
+      "loss": 1.4818,
+      "step": 16850
+    },
+    {
+      "epoch": 1.299900007691716,
+      "grad_norm": 2.497528553009033,
+      "learning_rate": 0.00011486897124899358,
+      "loss": 1.4176,
+      "step": 16900
+    },
+    {
+      "epoch": 1.3037458657026382,
+      "grad_norm": 1.6085398197174072,
+      "learning_rate": 0.00011460925122717712,
+      "loss": 1.4373,
+      "step": 16950
+    },
+    {
+      "epoch": 1.3075917237135606,
+      "grad_norm": 1.8937525749206543,
+      "learning_rate": 0.00011434953120536062,
+      "loss": 1.4823,
+      "step": 17000
+    },
+    {
+      "epoch": 1.3075917237135606,
+      "eval_loss": 1.4854488372802734,
+      "eval_runtime": 17.8761,
+      "eval_samples_per_second": 55.941,
+      "eval_steps_per_second": 13.985,
+      "step": 17000
     }
   ],
   "logging_steps": 50,