Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f32d05e67336f892684b8c1fd2d6e5abd7a19c7a3f8e60643ae3b6f25370775f
+oid sha256:49067541189efe886db6963256073252d295ca9bdc1760d32149b3d192043cff
 size 893438545

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:518db281c68ab44504d89afd4194f546b333ca0c39d6413cfd75d77184014cc2
+oid sha256:b40587551c6359d244432c6aa9fe4499a1dd75262644d31886b6f012e695e7e9
 size 449471589

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a8f9792121fbd04e7178f7b4e43756a5886a8af837b3120c6d50a7ef7909014
+oid sha256:4b9ef78655e6bb3bfb1e8e0536d4b8d1bed93f33651cb43025c88490b2ea3396
 size 15459

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61491770594419f1f65e70a85ff1045ef5f026004b37d4207f8d0e52435942a4
+oid sha256:a41408f29059abc49e36ed4313f4edd9d8787eb8db25f44722ca806febce9e97
 size 559

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a1bbe8751f81eef53f97fffae533ac792444b305682e60f93b951d5e4e28b33
+oid sha256:6df48bd3acfa1b41b6ce6c0df78ab20393bdc886bb13843361b0b4ecc2d1016d
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.3058239749281797,
-  "global_step": 15000,
+  "epoch": 1.7410986332375729,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -210,11 +210,79 @@
       "eval_samples_per_second": 54.906,
       "eval_steps_per_second": 0.868,
       "step": 15000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.550930216588928e-05,
+      "loss": 0.4482,
+      "step": 15500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.517085404351901e-05,
+      "loss": 0.4455,
+      "step": 16000
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.48225907448766e-05,
+      "loss": 0.4449,
+      "step": 16500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.44645981987198e-05,
+      "loss": 0.4454,
+      "step": 17000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.409696473435264e-05,
+      "loss": 0.4449,
+      "step": 17500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.371978105983142e-05,
+      "loss": 0.4436,
+      "step": 18000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.333314023958391e-05,
+      "loss": 0.4442,
+      "step": 18500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.293713767144707e-05,
+      "loss": 0.4429,
+      "step": 19000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.25326907757735e-05,
+      "loss": 0.4427,
+      "step": 19500
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.211827834753872e-05,
+      "loss": 0.4428,
+      "step": 20000
+    },
+    {
+      "epoch": 1.74,
+      "eval_loss": 0.4262467920780182,
+      "eval_runtime": 57.1744,
+      "eval_samples_per_second": 87.452,
+      "eval_steps_per_second": 1.382,
+      "step": 20000
     }
   ],
   "max_steps": 100000,
   "num_train_epochs": 9,
-  "total_flos": 7.064748208279152e+20,
+  "total_flos": 9.419678881926768e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:518db281c68ab44504d89afd4194f546b333ca0c39d6413cfd75d77184014cc2
+oid sha256:b40587551c6359d244432c6aa9fe4499a1dd75262644d31886b6f012e695e7e9
 size 449471589