Training in progress, step 26000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f097330090d54848c18e4bda429759e1abba010929a3213906365a68cd5e649a
+oid sha256:766f39cf2aa03b06ce0544465f23f42d796935dadd6619740980847fa379f296
 size 579824888

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fec1594771eede4505411cf0b3849475f8da63ebefb31deacc4aea691322fb69
+oid sha256:ad971091ff30bc232ebf11b75c27a84c1325701e13bc4a6246fcc2d26e336bb4
 size 1159794763

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12d604eeadf163b978dd48de387cbebeba84d52ce3a59a4bcf73ab714592e48f
+oid sha256:232768e06875e03d77948d18a567c6062bc64ee278f8d41937981a714e6a2d53
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4793a1af866198c083bf7fdbba00d5a4fb25389a6bed64d9ca675044e94ad0ae
+oid sha256:2c5343924a6852b4d8168211c95036da6881053e86461e5c21fd663c6b1f4a7b
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.48,
+  "epoch": 0.52,
   "eval_steps": 1000,
-  "global_step": 24000,
+  "global_step": 26000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1904,6 +1904,164 @@
       "eval_samples_per_second": 266.931,
       "eval_steps_per_second": 16.778,
       "step": 24000
+    },
+    {
+      "epoch": 0.482,
+      "grad_norm": 0.15001367032527924,
+      "learning_rate": 0.0005307348185995866,
+      "loss": 3.37049560546875,
+      "step": 24100
+    },
+    {
+      "epoch": 0.484,
+      "grad_norm": 0.15163348615169525,
+      "learning_rate": 0.0005275891420647307,
+      "loss": 3.3899249267578124,
+      "step": 24200
+    },
+    {
+      "epoch": 0.486,
+      "grad_norm": 0.13754494488239288,
+      "learning_rate": 0.0005244423697932499,
+      "loss": 3.397990417480469,
+      "step": 24300
+    },
+    {
+      "epoch": 0.488,
+      "grad_norm": 0.13546991348266602,
+      "learning_rate": 0.0005212946267630653,
+      "loss": 3.4154788208007814,
+      "step": 24400
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.17627890408039093,
+      "learning_rate": 0.0005181460379906534,
+      "loss": 3.4125030517578123,
+      "step": 24500
+    },
+    {
+      "epoch": 0.492,
+      "grad_norm": 0.14234718680381775,
+      "learning_rate": 0.0005149967285260802,
+      "loss": 3.4010299682617187,
+      "step": 24600
+    },
+    {
+      "epoch": 0.494,
+      "grad_norm": 0.1589398831129074,
+      "learning_rate": 0.0005118468234480345,
+      "loss": 3.4132962036132812,
+      "step": 24700
+    },
+    {
+      "epoch": 0.496,
+      "grad_norm": 0.12560983002185822,
+      "learning_rate": 0.0005086964478588614,
+      "loss": 3.3963751220703124,
+      "step": 24800
+    },
+    {
+      "epoch": 0.498,
+      "grad_norm": 0.14177259802818298,
+      "learning_rate": 0.0005055457268795923,
+      "loss": 3.3922305297851563,
+      "step": 24900
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.13485221564769745,
+      "learning_rate": 0.0005023947856449762,
+      "loss": 3.3956787109375,
+      "step": 25000
+    },
+    {
+      "epoch": 0.5,
+      "eval_accuracy": 0.3757247812423439,
+      "eval_loss": 3.3547720909118652,
+      "eval_runtime": 6.4257,
+      "eval_samples_per_second": 302.069,
+      "eval_steps_per_second": 18.986,
+      "step": 25000
+    },
+    {
+      "epoch": 0.502,
+      "grad_norm": 0.1296602189540863,
+      "learning_rate": 0.00049924374929851,
+      "loss": 3.4078851318359376,
+      "step": 25100
+    },
+    {
+      "epoch": 0.504,
+      "grad_norm": 0.17173421382904053,
+      "learning_rate": 0.0004960927429874685,
+      "loss": 3.378636779785156,
+      "step": 25200
+    },
+    {
+      "epoch": 0.506,
+      "grad_norm": 0.14787626266479492,
+      "learning_rate": 0.0004929418918579327,
+      "loss": 3.3894952392578124,
+      "step": 25300
+    },
+    {
+      "epoch": 0.508,
+      "grad_norm": 0.14193665981292725,
+      "learning_rate": 0.0004897913210498212,
+      "loss": 3.3784576416015626,
+      "step": 25400
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.34911948442459106,
+      "learning_rate": 0.0004866411556919189,
+      "loss": 3.4079013061523438,
+      "step": 25500
+    },
+    {
+      "epoch": 0.512,
+      "grad_norm": 0.14956624805927277,
+      "learning_rate": 0.00048349152089690765,
+      "loss": 3.395030517578125,
+      "step": 25600
+    },
+    {
+      "epoch": 0.514,
+      "grad_norm": 0.1397118866443634,
+      "learning_rate": 0.0004803425417563974,
+      "loss": 3.3780328369140626,
+      "step": 25700
+    },
+    {
+      "epoch": 0.516,
+      "grad_norm": 0.13487640023231506,
+      "learning_rate": 0.00047719434333595825,
+      "loss": 3.3928378295898436,
+      "step": 25800
+    },
+    {
+      "epoch": 0.518,
+      "grad_norm": 0.14225220680236816,
+      "learning_rate": 0.0004740470506701529,
+      "loss": 3.381436767578125,
+      "step": 25900
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.13680697977542877,
+      "learning_rate": 0.0004709007887575706,
+      "loss": 3.378739318847656,
+      "step": 26000
+    },
+    {
+      "epoch": 0.52,
+      "eval_accuracy": 0.3769195171452164,
+      "eval_loss": 3.345224142074585,
+      "eval_runtime": 6.5169,
+      "eval_samples_per_second": 297.843,
+      "eval_steps_per_second": 18.721,
+      "step": 26000
     }
   ],
   "logging_steps": 100,
@@ -1923,7 +2081,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 9.4079060803584e+17,
+  "total_flos": 1.01918982537216e+18,
   "train_batch_size": 120,
   "trial_name": null,
   "trial_params": null