Training in progress, step 17000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c79bd6abfc765e48344f1eaf2bac6b19b1f2185bd8d532b6430edddeb3cf598
+oid sha256:1d163d6a9ddc974e2180a0b34dccdebf25df2e06b238d605670c84471509b1d7
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57817e37c2bede420a10ad726002c513c322991defcb7b6d376b6f78c40cfde6
+oid sha256:48e55dadd5444de13311e56742ff892d8682c557f87064412817be365756d8c7
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d74c3606ba0286a1d15828b59a77b9982057d18434b39f6b0feddf0ba035b2fb
+oid sha256:fc77b7658a34f923b1da47ba7dbd2fd74f55df477ff11a70a0b6f32488a49438
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a861053ff17e2abedc4d7c6ad9da2b90a9c5355f647fb00b85fde4046e6f77e9
+oid sha256:a815a54c3da05d513f8fc01a4362c2083ab81366f45a37e94100e41551fa2292
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b186621336b0abea983113004ecd9b3118c0bc11e75a3da16b0bb6e48f4f1d5
+oid sha256:6862466a821eb915ca4f2b1057c449dfe2dac434c155b9783db0cf817969e1cf
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e7ce0d53f1187963494f7c07c00eb523b42730ba00d8fde1d32c14564031cf5
+oid sha256:3597272c85a71f80afe5d0d108e62f0719dbbc66ca2fad8b5c08fb0ab04b2895
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 16500,
-  "best_metric": 0.8012632131576538,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-16500",
-  "epoch": 1.2691331436043383,
+  "best_global_step": 17000,
+  "best_metric": 0.8002509474754333,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-17000",
+  "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": 55.937,
       "eval_steps_per_second": 13.984,
       "step": 16500
+    },
+    {
+      "epoch": 1.2729790016152602,
+      "grad_norm": 0.7161915302276611,
+      "learning_rate": 0.00011667662260083629,
+      "loss": 0.7969,
+      "step": 16550
+    },
+    {
+      "epoch": 1.2768248596261826,
+      "grad_norm": 0.7693639397621155,
+      "learning_rate": 0.00011641690257901983,
+      "loss": 0.7909,
+      "step": 16600
+    },
+    {
+      "epoch": 1.2806707176371048,
+      "grad_norm": 0.6645983457565308,
+      "learning_rate": 0.00011615718255720334,
+      "loss": 0.7816,
+      "step": 16650
+    },
+    {
+      "epoch": 1.284516575648027,
+      "grad_norm": 0.7736928462982178,
+      "learning_rate": 0.00011589746253538687,
+      "loss": 0.745,
+      "step": 16700
+    },
+    {
+      "epoch": 1.2883624336589494,
+      "grad_norm": 0.6743229627609253,
+      "learning_rate": 0.00011563774251357038,
+      "loss": 0.7856,
+      "step": 16750
+    },
+    {
+      "epoch": 1.2883624336589494,
+      "eval_loss": 0.8006194233894348,
+      "eval_runtime": 17.9046,
+      "eval_samples_per_second": 55.852,
+      "eval_steps_per_second": 13.963,
+      "step": 16750
+    },
+    {
+      "epoch": 1.2922082916698716,
+      "grad_norm": 0.8146863579750061,
+      "learning_rate": 0.0001153780224917539,
+      "loss": 0.7559,
+      "step": 16800
+    },
+    {
+      "epoch": 1.2960541496807938,
+      "grad_norm": 0.9186645746231079,
+      "learning_rate": 0.00011511830246993742,
+      "loss": 0.7902,
+      "step": 16850
+    },
+    {
+      "epoch": 1.299900007691716,
+      "grad_norm": 0.7634202241897583,
+      "learning_rate": 0.00011485858244812093,
+      "loss": 0.749,
+      "step": 16900
+    },
+    {
+      "epoch": 1.3037458657026382,
+      "grad_norm": 0.890457272529602,
+      "learning_rate": 0.00011459886242630445,
+      "loss": 0.7539,
+      "step": 16950
+    },
+    {
+      "epoch": 1.3075917237135606,
+      "grad_norm": 0.8466306328773499,
+      "learning_rate": 0.00011433914240448797,
+      "loss": 0.7997,
+      "step": 17000
+    },
+    {
+      "epoch": 1.3075917237135606,
+      "eval_loss": 0.8002509474754333,
+      "eval_runtime": 17.9496,
+      "eval_samples_per_second": 55.712,
+      "eval_steps_per_second": 13.928,
+      "step": 17000
     }
   ],
   "logging_steps": 50,