Training in progress, step 7500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:408c18ba3992e03f34e827417b7daf252b66eb9ab444c173edf10367465681b5
+oid sha256:b15dd6962d86db50650c0eddcefd00dea38eda781c7e3e209372344783b063ab
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d7aa88aa0d1d0ec7fe3d76e4e8d27243c65037e1872b868b51819cd8e66d8e7
+oid sha256:928b3e8998ba045a99767c2828895347f34a6cf92e03f46714188bc3addda823
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3460a162be2eb8933a08c339a02a822773fed1405794db28d2550dafdac12db
+oid sha256:a3e2cd559a12cb39d13f548f421501a93ca642c8a0b1f87792aa4f56c33af4af
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f4acfd5fa4964d8f20ab734945eda1f469ecfbbe259a3fd74f731ec4baabed1
+oid sha256:1d18ab5183ebf11753d526532b8569bec6f06614d4e88fe8e90cb6ae5b6d98c0
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:021615b61b72c4c82d5fefd973bcdc47aefa683f82add6615dce2f0c0d108036
+oid sha256:1e3801f627da5685ffd4894012426c784cb58a2cca4cc7d65b5ce5beaa645a63
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4c7b441125154cea489ca7028527a0b9f859381d9d7a1888e04b52f3f2bc3c5
+oid sha256:c3b026101061ccf53c244a631f5dc343dee66fbeba0e3ca39e731be93c209c38
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 7000,
-  "best_metric": 4.429446220397949,
+  "best_global_step": 7250,
+  "best_metric": 4.42822265625,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-7000",
-  "epoch": 0.5384201215291131,
+  "epoch": 0.5768787016383355,
   "eval_steps": 250,
-  "global_step": 7000,
+  "global_step": 7500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1212,6 +1212,92 @@
       "eval_samples_per_second": 53.664,
       "eval_steps_per_second": 13.416,
       "step": 7000
+    },
+    {
+      "epoch": 0.5422659795400354,
+      "grad_norm": 6.055607795715332,
+      "learning_rate": 0.00016602862114640417,
+      "loss": 4.3805,
+      "step": 7050
+    },
+    {
+      "epoch": 0.5461118375509576,
+      "grad_norm": 3.190605878829956,
+      "learning_rate": 0.00016576890112458768,
+      "loss": 4.561,
+      "step": 7100
+    },
+    {
+      "epoch": 0.5499576955618799,
+      "grad_norm": 2.95857834815979,
+      "learning_rate": 0.00016550918110277122,
+      "loss": 4.5957,
+      "step": 7150
+    },
+    {
+      "epoch": 0.5538035535728021,
+      "grad_norm": 5.055838108062744,
+      "learning_rate": 0.00016524946108095474,
+      "loss": 4.3916,
+      "step": 7200
+    },
+    {
+      "epoch": 0.5576494115837243,
+      "grad_norm": 6.25083589553833,
+      "learning_rate": 0.00016498974105913825,
+      "loss": 4.6649,
+      "step": 7250
+    },
+    {
+      "epoch": 0.5576494115837243,
+      "eval_loss": 4.42822265625,
+      "eval_runtime": 18.5876,
+      "eval_samples_per_second": 53.799,
+      "eval_steps_per_second": 13.45,
+      "step": 7250
+    },
+    {
+      "epoch": 0.5614952695946466,
+      "grad_norm": 3.6657636165618896,
+      "learning_rate": 0.0001647300210373218,
+      "loss": 4.5059,
+      "step": 7300
+    },
+    {
+      "epoch": 0.5653411276055688,
+      "grad_norm": 3.6100645065307617,
+      "learning_rate": 0.0001644703010155053,
+      "loss": 4.4379,
+      "step": 7350
+    },
+    {
+      "epoch": 0.5691869856164911,
+      "grad_norm": 3.535804271697998,
+      "learning_rate": 0.0001642105809936888,
+      "loss": 4.3881,
+      "step": 7400
+    },
+    {
+      "epoch": 0.5730328436274132,
+      "grad_norm": 2.9636013507843018,
+      "learning_rate": 0.00016395086097187233,
+      "loss": 4.491,
+      "step": 7450
+    },
+    {
+      "epoch": 0.5768787016383355,
+      "grad_norm": 2.9023678302764893,
+      "learning_rate": 0.00016369114095005584,
+      "loss": 4.5567,
+      "step": 7500
+    },
+    {
+      "epoch": 0.5768787016383355,
+      "eval_loss": 4.459234237670898,
+      "eval_runtime": 18.5868,
+      "eval_samples_per_second": 53.802,
+      "eval_steps_per_second": 13.45,
+      "step": 7500
     }
   ],
   "logging_steps": 50,