Training in progress, step 3000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e10a7c8bb0d516328a4f08fb821bc0e53102e1b0a0d8ea1d6c8cc3bce9dca90
+oid sha256:84adf5fb4a3947b35993a790cf4973ef5593a87e6e164d9578f03b8b9c27e099
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdb38bb607b4da045b51a34cfbf7428a4bf481a4b7dcfcd083d72647f444c879
+oid sha256:308f424d188d86d1fcd72962eb2fa00bace74a0cc7f5cd9ae44b93fa7b48ebe5
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d026d37c2efc698996a40494d349ec54a4ff40c81599ec63f15410e52a0e87f
+oid sha256:65a04555108b26fbf193a851c8c842cf49a4c844bc1961b985cfc83f43bcde61
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:100aab8649e7180e1ab3ef8feaab197f99e31e9f728f5189c5431d80dbc2eacd
-size 14244
+oid sha256:238412d43efe8b606030d3301ede6818fb1a62974cd0cbb655e125124222d862
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:996af29fdb07e722f1f70dd9ba54354e5a4baf4f673aef8bef5139d74d107a73
+oid sha256:04b1366b3719985cdc1dc0eb0fa220f7b35ea714d4ff368885830073c629f4d9
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d57c57de49427b666f955c0cc0173da1ca9651de809c40ac268e017db320996
+oid sha256:2976163d4a5cd674bbf6d82ca442cb1ceba9d5f9d444c557ab914dab5858c6ce
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 2500,
-  "best_metric": 1.5644338130950928,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-2500",
-  "epoch": 0.19229290054611184,
+  "best_global_step": 3000,
+  "best_metric": 1.5423797369003296,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-3000",
+  "epoch": 0.2307514806553342,
   "eval_steps": 250,
-  "global_step": 2500,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -438,6 +438,92 @@
       "eval_samples_per_second": 59.426,
       "eval_steps_per_second": 14.856,
       "step": 2500
+    },
+    {
+      "epoch": 0.19613875855703408,
+      "grad_norm": 0.8316934704780579,
+      "learning_rate": 0.00018937745110770592,
+      "loss": 1.5182,
+      "step": 2550
+    },
+    {
+      "epoch": 0.19998461656795632,
+      "grad_norm": 1.3362255096435547,
+      "learning_rate": 0.0001891177310858894,
+      "loss": 1.5557,
+      "step": 2600
+    },
+    {
+      "epoch": 0.20383047457887854,
+      "grad_norm": 1.2854304313659668,
+      "learning_rate": 0.00018885801106407294,
+      "loss": 1.5417,
+      "step": 2650
+    },
+    {
+      "epoch": 0.20767633258980078,
+      "grad_norm": 1.7450816631317139,
+      "learning_rate": 0.00018859829104225646,
+      "loss": 1.5489,
+      "step": 2700
+    },
+    {
+      "epoch": 0.21152219060072303,
+      "grad_norm": 1.546242117881775,
+      "learning_rate": 0.00018833857102043997,
+      "loss": 1.4724,
+      "step": 2750
+    },
+    {
+      "epoch": 0.21152219060072303,
+      "eval_loss": 1.5916800498962402,
+      "eval_runtime": 16.9148,
+      "eval_samples_per_second": 59.12,
+      "eval_steps_per_second": 14.78,
+      "step": 2750
+    },
+    {
+      "epoch": 0.21536804861164527,
+      "grad_norm": 1.8118102550506592,
+      "learning_rate": 0.0001880788509986235,
+      "loss": 1.511,
+      "step": 2800
+    },
+    {
+      "epoch": 0.21921390662256748,
+      "grad_norm": 1.7519344091415405,
+      "learning_rate": 0.00018781913097680702,
+      "loss": 1.5161,
+      "step": 2850
+    },
+    {
+      "epoch": 0.22305976463348973,
+      "grad_norm": 1.7779595851898193,
+      "learning_rate": 0.00018755941095499053,
+      "loss": 1.6146,
+      "step": 2900
+    },
+    {
+      "epoch": 0.22690562264441197,
+      "grad_norm": 1.0836044549942017,
+      "learning_rate": 0.00018729969093317405,
+      "loss": 1.5074,
+      "step": 2950
+    },
+    {
+      "epoch": 0.2307514806553342,
+      "grad_norm": 1.4328197240829468,
+      "learning_rate": 0.00018703997091135756,
+      "loss": 1.5865,
+      "step": 3000
+    },
+    {
+      "epoch": 0.2307514806553342,
+      "eval_loss": 1.5423797369003296,
+      "eval_runtime": 16.9029,
+      "eval_samples_per_second": 59.161,
+      "eval_steps_per_second": 14.79,
+      "step": 3000
     }
   ],
   "logging_steps": 50,