Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6839e8fd9262ae9437bdd437abd1bad3c3b82511739449fec8292a69e4d5b10b
+oid sha256:98bc95b3264bd457f2f822f8bff3c25921dcd820b2ab11607b18dca44e5ccdc8
 size 893438545

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b31e0a533670bdb77786399adaddbe18894c2044c85487dcc9d0c41d55182f57
+oid sha256:65b4f4d147bd2f30b9ef842bfdf3fee1e6165115164d8a1391c9c747271d8965
 size 449471589

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5d4271a4bb1cdb3db9b1359f873030d4eefa7f15660df68f0302617274cecfd
+oid sha256:21343af8626647d218ed09443d9afdff2e884071902e842787b88cc7208f1f68
 size 15587

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5a758f6bb9e87b722e928f89b126a9e2bb6d7d2276a24ffb9fee6ce69354628
+oid sha256:bfd5a73d368b8347a9555d7b94e42b0bfbb251b0b8d4142bfa4cf49d9c2e9ffd
 size 559

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a1bbe8751f81eef53f97fffae533ac792444b305682e60f93b951d5e4e28b33
+oid sha256:a6464ca3b88e496d5d945a7955ac6cabd3c94f9611fe337e2e8a9af945aa679a
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.305937663242208,
-  "global_step": 15000,
+  "epoch": 1.7412502176562772,
+  "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": 293.963,
       "eval_steps_per_second": 4.645,
       "step": 15000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.550863512110018e-05,
+      "loss": 0.4464,
+      "step": 15500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.517016728422667e-05,
+      "loss": 0.4457,
+      "step": 16000
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.482188444052858e-05,
+      "loss": 0.4437,
+      "step": 16500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.446387252358614e-05,
+      "loss": 0.4443,
+      "step": 17000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.40962198674828e-05,
+      "loss": 0.4441,
+      "step": 17500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.371901718501017e-05,
+      "loss": 0.4438,
+      "step": 18000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.333314023958391e-05,
+      "loss": 0.4421,
+      "step": 18500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.293713767144707e-05,
+      "loss": 0.4415,
+      "step": 19000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.253187106312908e-05,
+      "loss": 0.442,
+      "step": 19500
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.211744040810141e-05,
+      "loss": 0.441,
+      "step": 20000
+    },
+    {
+      "epoch": 1.74,
+      "eval_loss": 0.4204435646533966,
+      "eval_runtime": 18.5947,
+      "eval_samples_per_second": 268.894,
+      "eval_steps_per_second": 4.249,
+      "step": 20000
     }
   ],
   "max_steps": 100000,
   "num_train_epochs": 9,
-  "total_flos": 7.064605817122141e+20,
+  "total_flos": 9.419536490769757e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b31e0a533670bdb77786399adaddbe18894c2044c85487dcc9d0c41d55182f57
+oid sha256:65b4f4d147bd2f30b9ef842bfdf3fee1e6165115164d8a1391c9c747271d8965
 size 449471589