Training in progress, step 3500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84adf5fb4a3947b35993a790cf4973ef5593a87e6e164d9578f03b8b9c27e099
+oid sha256:11a062f0d5247b9505627696a295b4feca04c6c45dd688fcbc2c07d7828e414c
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:308f424d188d86d1fcd72962eb2fa00bace74a0cc7f5cd9ae44b93fa7b48ebe5
+oid sha256:1ce273710431703c604c9d640c23434ce3f2b038664d7b3df1e4a13e933764d0
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65a04555108b26fbf193a851c8c842cf49a4c844bc1961b985cfc83f43bcde61
+oid sha256:eb347b9f0c8f6a0d9a0e6b08949ae99d2459f1f662e3a7dd16f39b8ba0ddf69c
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:238412d43efe8b606030d3301ede6818fb1a62974cd0cbb655e125124222d862
-size 14180
+oid sha256:3cf21f101c997b0643b73ca8f669be23f0abb9e5a2c6fedb07e17298d9cdd268
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04b1366b3719985cdc1dc0eb0fa220f7b35ea714d4ff368885830073c629f4d9
+oid sha256:028b7d42ce0dbafb2a0126ad830dd1957166a4ca85043c6644487cae8315bfe9
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2976163d4a5cd674bbf6d82ca442cb1ceba9d5f9d444c557ab914dab5858c6ce
+oid sha256:d16b13e1d35412f9085034fa2481674486c7b966c61cae81848c36587b827f3d
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 3000,
-  "best_metric": 1.5423797369003296,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-3000",
-  "epoch": 0.2307514806553342,
+  "best_global_step": 3500,
+  "best_metric": 1.5354665517807007,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-3500",
+  "epoch": 0.26921006076455656,
   "eval_steps": 250,
-  "global_step": 3000,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -524,6 +524,92 @@
       "eval_samples_per_second": 59.161,
       "eval_steps_per_second": 14.79,
       "step": 3000
+    },
+    {
+      "epoch": 0.23459733866625646,
+      "grad_norm": 3.606706142425537,
+      "learning_rate": 0.00018678025088954107,
+      "loss": 1.5998,
+      "step": 3050
+    },
+    {
+      "epoch": 0.23844319667717867,
+      "grad_norm": 1.2053996324539185,
+      "learning_rate": 0.0001865205308677246,
+      "loss": 1.5512,
+      "step": 3100
+    },
+    {
+      "epoch": 0.2422890546881009,
+      "grad_norm": 1.5675193071365356,
+      "learning_rate": 0.00018626081084590813,
+      "loss": 1.5495,
+      "step": 3150
+    },
+    {
+      "epoch": 0.24613491269902316,
+      "grad_norm": 1.1827441453933716,
+      "learning_rate": 0.00018600109082409164,
+      "loss": 1.6509,
+      "step": 3200
+    },
+    {
+      "epoch": 0.2499807707099454,
+      "grad_norm": 1.7992998361587524,
+      "learning_rate": 0.00018574137080227515,
+      "loss": 1.5617,
+      "step": 3250
+    },
+    {
+      "epoch": 0.2499807707099454,
+      "eval_loss": 1.5403118133544922,
+      "eval_runtime": 16.9627,
+      "eval_samples_per_second": 58.953,
+      "eval_steps_per_second": 14.738,
+      "step": 3250
+    },
+    {
+      "epoch": 0.25382662872086764,
+      "grad_norm": 1.3044147491455078,
+      "learning_rate": 0.00018548165078045866,
+      "loss": 1.5226,
+      "step": 3300
+    },
+    {
+      "epoch": 0.2576724867317899,
+      "grad_norm": 1.4057557582855225,
+      "learning_rate": 0.0001852219307586422,
+      "loss": 1.517,
+      "step": 3350
+    },
+    {
+      "epoch": 0.26151834474271207,
+      "grad_norm": 0.9969176650047302,
+      "learning_rate": 0.00018496221073682572,
+      "loss": 1.5157,
+      "step": 3400
+    },
+    {
+      "epoch": 0.2653642027536343,
+      "grad_norm": 1.0183119773864746,
+      "learning_rate": 0.00018470249071500923,
+      "loss": 1.5012,
+      "step": 3450
+    },
+    {
+      "epoch": 0.26921006076455656,
+      "grad_norm": 1.1896519660949707,
+      "learning_rate": 0.00018444277069319277,
+      "loss": 1.4655,
+      "step": 3500
+    },
+    {
+      "epoch": 0.26921006076455656,
+      "eval_loss": 1.5354665517807007,
+      "eval_runtime": 16.8964,
+      "eval_samples_per_second": 59.184,
+      "eval_steps_per_second": 14.796,
+      "step": 3500
     }
   ],
   "logging_steps": 50,