Training in progress, step 13000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a623f2e366d56a653f11e3f01819889dc1645d4d1136c5b510bf0c1c92dab66
+oid sha256:34351883b0f6e42fd271ed7ebc5095493f25642a1a75cc341d7789a6aeca36d0
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6015f4fec963cdcdbd9534ce342304afb8094ad5f761558a674ddf3068ab0335
+oid sha256:39353fd13db5daeeb97a98bc4902673ea2228d2bc5b271b25f3916da657ebef6
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f314be3a1ba2ce8f5f725cf645fa00681145b061c07cf8ad59e1e13356924cb
+oid sha256:8448ab33e8a6201d670f6c9fb2d20010c66be0d08d4deb6a1a6fb49d8b9c406d
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:082b149e63f0f47e9e60d37040a9d015e2fa6a3d80388f6fa7d859aa3870fd21
-size 14308
+oid sha256:9893f35eae5cc019e9b33dde5eb3fb2bd041941900e50675b17b7d1262060d8f
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0678e6da71fe39cdc83977acb14900d45184a734b31b0b8a20ead5262367273c
+oid sha256:e2b1ebe381b38de043f6835dd6952f12b83db6f06939f85a5aefa0eb3268de14
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7680f728f95fcd934cd32ef342f02b8f2b507472a27e9e8de734cc5468e1428
+oid sha256:adb9317783409b5e27bc46a88956d26ababa0364a288c25ea22c620413a62bd0
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 12500,
-  "best_metric": -30.455406188964844,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-12500",
-  "epoch": 0.9614645027305592,
+  "best_global_step": 13000,
+  "best_metric": -30.469120025634766,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-13000",
+  "epoch": 0.9999230828397816,
   "eval_steps": 250,
-  "global_step": 12500,
+  "global_step": 13000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2158,6 +2158,92 @@
       "eval_samples_per_second": 61.423,
       "eval_steps_per_second": 15.356,
       "step": 12500
+    },
+    {
+      "epoch": 0.9653103607414815,
+      "grad_norm": 0.1690986603498459,
+      "learning_rate": 0.00013744902994571853,
+      "loss": -30.4565,
+      "step": 12550
+    },
+    {
+      "epoch": 0.9691562187524037,
+      "grad_norm": 0.15991179645061493,
+      "learning_rate": 0.00013718930992390204,
+      "loss": -30.4596,
+      "step": 12600
+    },
+    {
+      "epoch": 0.973002076763326,
+      "grad_norm": 0.3283814787864685,
+      "learning_rate": 0.00013692958990208556,
+      "loss": -30.4609,
+      "step": 12650
+    },
+    {
+      "epoch": 0.9768479347742481,
+      "grad_norm": 0.09776933491230011,
+      "learning_rate": 0.0001366698698802691,
+      "loss": -30.4628,
+      "step": 12700
+    },
+    {
+      "epoch": 0.9806937927851703,
+      "grad_norm": 0.16940400004386902,
+      "learning_rate": 0.0001364101498584526,
+      "loss": -30.4646,
+      "step": 12750
+    },
+    {
+      "epoch": 0.9806937927851703,
+      "eval_loss": -30.46562385559082,
+      "eval_runtime": 16.3759,
+      "eval_samples_per_second": 61.065,
+      "eval_steps_per_second": 15.266,
+      "step": 12750
+    },
+    {
+      "epoch": 0.9845396507960926,
+      "grad_norm": 0.1123572438955307,
+      "learning_rate": 0.0001361504298366361,
+      "loss": -30.4657,
+      "step": 12800
+    },
+    {
+      "epoch": 0.9883855088070148,
+      "grad_norm": 0.19518135488033295,
+      "learning_rate": 0.00013589070981481963,
+      "loss": -30.4662,
+      "step": 12850
+    },
+    {
+      "epoch": 0.9922313668179371,
+      "grad_norm": 0.09862416982650757,
+      "learning_rate": 0.00013563098979300315,
+      "loss": -30.4669,
+      "step": 12900
+    },
+    {
+      "epoch": 0.9960772248288593,
+      "grad_norm": 0.08367747068405151,
+      "learning_rate": 0.00013537126977118666,
+      "loss": -30.4679,
+      "step": 12950
+    },
+    {
+      "epoch": 0.9999230828397816,
+      "grad_norm": 0.110712930560112,
+      "learning_rate": 0.0001351115497493702,
+      "loss": -30.4689,
+      "step": 13000
+    },
+    {
+      "epoch": 0.9999230828397816,
+      "eval_loss": -30.469120025634766,
+      "eval_runtime": 16.2879,
+      "eval_samples_per_second": 61.395,
+      "eval_steps_per_second": 15.349,
+      "step": 13000
     }
   ],
   "logging_steps": 50,