Training in progress, step 7000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df74840f90de2e67e9907cab52791a6dd441f2b7df4dee24f3d27614fb7c4e81
+oid sha256:408c18ba3992e03f34e827417b7daf252b66eb9ab444c173edf10367465681b5
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35f6e0128f9fd6473636f2b7fa321fba2a34670190a7fcccacc9997f0e6e3efd
+oid sha256:0d7aa88aa0d1d0ec7fe3d76e4e8d27243c65037e1872b868b51819cd8e66d8e7
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e91da0cfbee4e888038757b7189ea413722951a4530435531acf1f53d94aedfd
+oid sha256:a3460a162be2eb8933a08c339a02a822773fed1405794db28d2550dafdac12db
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44c02dc0791054ba042a754ce4b193305b8c58f3c6ec32c6d59f64ae1e19274a
+oid sha256:2f4acfd5fa4964d8f20ab734945eda1f469ecfbbe259a3fd74f731ec4baabed1
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96851dceb7a310a70f7ecfde20fb5205a7253860759d2b8ce0924be746e3091f
+oid sha256:021615b61b72c4c82d5fefd973bcdc47aefa683f82add6615dce2f0c0d108036
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18b7addcef54bd827fd6d26a7fd7e8fd6d6f1182cd34f05f02a7c070592639be
+oid sha256:e4c7b441125154cea489ca7028527a0b9f859381d9d7a1888e04b52f3f2bc3c5
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 5000,
-  "best_metric": 4.465761661529541,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-5000",
-  "epoch": 0.4999615414198908,
+  "best_global_step": 7000,
+  "best_metric": 4.429446220397949,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-7000",
+  "epoch": 0.5384201215291131,
   "eval_steps": 250,
-  "global_step": 6500,
+  "global_step": 7000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1126,6 +1126,92 @@
       "eval_samples_per_second": 53.723,
       "eval_steps_per_second": 13.431,
       "step": 6500
+    },
+    {
+      "epoch": 0.503807399430813,
+      "grad_norm": 3.325056791305542,
+      "learning_rate": 0.00016861543256369634,
+      "loss": 4.618,
+      "step": 6550
+    },
+    {
+      "epoch": 0.5076532574417353,
+      "grad_norm": 4.026259899139404,
+      "learning_rate": 0.0001683609069423162,
+      "loss": 4.5485,
+      "step": 6600
+    },
+    {
+      "epoch": 0.5114991154526575,
+      "grad_norm": 3.1270413398742676,
+      "learning_rate": 0.00016810118692049973,
+      "loss": 4.4426,
+      "step": 6650
+    },
+    {
+      "epoch": 0.5153449734635798,
+      "grad_norm": 5.264435768127441,
+      "learning_rate": 0.00016784146689868324,
+      "loss": 4.3544,
+      "step": 6700
+    },
+    {
+      "epoch": 0.519190831474502,
+      "grad_norm": 4.349465847015381,
+      "learning_rate": 0.00016758694127730307,
+      "loss": 4.5912,
+      "step": 6750
+    },
+    {
+      "epoch": 0.519190831474502,
+      "eval_loss": 4.503914833068848,
+      "eval_runtime": 18.5932,
+      "eval_samples_per_second": 53.783,
+      "eval_steps_per_second": 13.446,
+      "step": 6750
+    },
+    {
+      "epoch": 0.5230366894854241,
+      "grad_norm": 2.5264992713928223,
+      "learning_rate": 0.00016732722125548658,
+      "loss": 4.3703,
+      "step": 6800
+    },
+    {
+      "epoch": 0.5268825474963464,
+      "grad_norm": 4.934820175170898,
+      "learning_rate": 0.00016706750123367012,
+      "loss": 4.5652,
+      "step": 6850
+    },
+    {
+      "epoch": 0.5307284055072686,
+      "grad_norm": 4.21425724029541,
+      "learning_rate": 0.00016680778121185363,
+      "loss": 4.4684,
+      "step": 6900
+    },
+    {
+      "epoch": 0.5345742635181909,
+      "grad_norm": 5.111146926879883,
+      "learning_rate": 0.00016654806119003715,
+      "loss": 4.5119,
+      "step": 6950
+    },
+    {
+      "epoch": 0.5384201215291131,
+      "grad_norm": 4.563775062561035,
+      "learning_rate": 0.00016628834116822069,
+      "loss": 4.489,
+      "step": 7000
+    },
+    {
+      "epoch": 0.5384201215291131,
+      "eval_loss": 4.429446220397949,
+      "eval_runtime": 18.6345,
+      "eval_samples_per_second": 53.664,
+      "eval_steps_per_second": 13.416,
+      "step": 7000
     }
   ],
   "logging_steps": 50,