Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f328e0037b7e952e07e4644c4009369ed6a6d95e4b94a53ca1d670f05f0897d5
+oid sha256:232b68b8ddafb01df2984a10e33488829130b30e041e48d066b54939473e6ff6
 size 891558696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:519af868d035d200ad2ccfac2e02cf6b17cf980e3bcb23ff5f68ab2f729de630
+oid sha256:7069a14fcc9bff0c6bc9038483cdefb26decf1998e8d418211dfbfa3e4a8af5e
 size 1783272762

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a29ed82ba21af44a36d8e59405dc8d62a827ba1626cee944f74dd401606bce37
+oid sha256:e79bcfaf683538279e39272f16342dfc788f3ded86099a7f67e3c0cbc2e642c8
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81759665abac97fd722dc6d155c208e433da3f7352b6dfc928e60adc595c0aee
+oid sha256:cfecd979e55961cee0575d5f79cb68e42062200bea1d0b301500013e33f96097
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.2369847148656845,
-  "best_model_checkpoint": "./fine-tuned/checkpoint-1500",
-  "epoch": 2.493765586034913,
+  "best_metric": 0.23246362805366516,
+  "best_model_checkpoint": "./fine-tuned/checkpoint-2000",
+  "epoch": 3.3250207813798838,
   "eval_steps": 100,
-  "global_step": 1500,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -25,9 +25,9 @@
     {
       "epoch": 0.1662510390689942,
       "eval_loss": 0.35283052921295166,
-      "eval_runtime": 38.0523,
-      "eval_samples_per_second": 12.982,
-      "eval_steps_per_second": 1.629,
+      "eval_runtime": 36.3275,
+      "eval_samples_per_second": 13.598,
+      "eval_steps_per_second": 1.707,
       "step": 100
     },
     {
@@ -47,9 +47,9 @@
     {
       "epoch": 0.3325020781379884,
       "eval_loss": 0.30865946412086487,
-      "eval_runtime": 38.2162,
-      "eval_samples_per_second": 12.926,
-      "eval_steps_per_second": 1.622,
+      "eval_runtime": 36.5834,
+      "eval_samples_per_second": 13.503,
+      "eval_steps_per_second": 1.695,
       "step": 200
     },
     {
@@ -69,9 +69,9 @@
     {
       "epoch": 0.49875311720698257,
       "eval_loss": 0.2891499996185303,
-      "eval_runtime": 38.1858,
-      "eval_samples_per_second": 12.937,
-      "eval_steps_per_second": 1.624,
+      "eval_runtime": 36.4356,
+      "eval_samples_per_second": 13.558,
+      "eval_steps_per_second": 1.702,
       "step": 300
     },
     {
@@ -91,9 +91,9 @@
     {
       "epoch": 0.6650041562759768,
       "eval_loss": 0.2757515609264374,
-      "eval_runtime": 38.2247,
-      "eval_samples_per_second": 12.924,
-      "eval_steps_per_second": 1.622,
+      "eval_runtime": 36.561,
+      "eval_samples_per_second": 13.512,
+      "eval_steps_per_second": 1.696,
       "step": 400
     },
     {
@@ -113,9 +113,9 @@
     {
       "epoch": 0.8312551953449709,
       "eval_loss": 0.2673098146915436,
-      "eval_runtime": 38.0207,
-      "eval_samples_per_second": 12.993,
-      "eval_steps_per_second": 1.631,
+      "eval_runtime": 36.4611,
+      "eval_samples_per_second": 13.549,
+      "eval_steps_per_second": 1.7,
       "step": 500
     },
     {
@@ -135,9 +135,9 @@
     {
       "epoch": 0.9975062344139651,
       "eval_loss": 0.26070085167884827,
-      "eval_runtime": 37.7694,
-      "eval_samples_per_second": 13.079,
-      "eval_steps_per_second": 1.642,
+      "eval_runtime": 36.5977,
+      "eval_samples_per_second": 13.498,
+      "eval_steps_per_second": 1.694,
       "step": 600
     },
     {
@@ -157,9 +157,9 @@
     {
       "epoch": 1.1637572734829593,
       "eval_loss": 0.256939560174942,
-      "eval_runtime": 38.0385,
-      "eval_samples_per_second": 12.987,
-      "eval_steps_per_second": 1.63,
+      "eval_runtime": 36.5571,
+      "eval_samples_per_second": 13.513,
+      "eval_steps_per_second": 1.696,
       "step": 700
     },
     {
@@ -179,9 +179,9 @@
     {
       "epoch": 1.3300083125519535,
       "eval_loss": 0.2525966763496399,
-      "eval_runtime": 38.1831,
-      "eval_samples_per_second": 12.938,
-      "eval_steps_per_second": 1.624,
+      "eval_runtime": 36.5366,
+      "eval_samples_per_second": 13.521,
+      "eval_steps_per_second": 1.697,
       "step": 800
     },
     {
@@ -201,9 +201,9 @@
     {
       "epoch": 1.4962593516209477,
       "eval_loss": 0.24994711577892303,
-      "eval_runtime": 37.4678,
-      "eval_samples_per_second": 13.185,
-      "eval_steps_per_second": 1.655,
+      "eval_runtime": 36.5909,
+      "eval_samples_per_second": 13.501,
+      "eval_steps_per_second": 1.694,
       "step": 900
     },
     {
@@ -223,9 +223,9 @@
     {
       "epoch": 1.6625103906899419,
       "eval_loss": 0.246443971991539,
-      "eval_runtime": 37.8324,
-      "eval_samples_per_second": 13.058,
-      "eval_steps_per_second": 1.639,
+      "eval_runtime": 36.514,
+      "eval_samples_per_second": 13.529,
+      "eval_steps_per_second": 1.698,
       "step": 1000
     },
     {
@@ -245,9 +245,9 @@
     {
       "epoch": 1.828761429758936,
       "eval_loss": 0.24409395456314087,
-      "eval_runtime": 37.7821,
-      "eval_samples_per_second": 13.075,
-      "eval_steps_per_second": 1.641,
+      "eval_runtime": 36.5322,
+      "eval_samples_per_second": 13.522,
+      "eval_steps_per_second": 1.697,
       "step": 1100
     },
     {
@@ -267,9 +267,9 @@
     {
       "epoch": 1.9950124688279303,
       "eval_loss": 0.2411041557788849,
-      "eval_runtime": 37.3119,
-      "eval_samples_per_second": 13.24,
-      "eval_steps_per_second": 1.662,
+      "eval_runtime": 36.4594,
+      "eval_samples_per_second": 13.549,
+      "eval_steps_per_second": 1.701,
       "step": 1200
     },
     {
@@ -289,9 +289,9 @@
     {
       "epoch": 2.1612635078969245,
       "eval_loss": 0.23963774740695953,
-      "eval_runtime": 37.7867,
-      "eval_samples_per_second": 13.073,
-      "eval_steps_per_second": 1.641,
+      "eval_runtime": 36.6761,
+      "eval_samples_per_second": 13.469,
+      "eval_steps_per_second": 1.69,
       "step": 1300
     },
     {
@@ -311,9 +311,9 @@
     {
       "epoch": 2.3275145469659186,
       "eval_loss": 0.23829442262649536,
-      "eval_runtime": 37.7411,
-      "eval_samples_per_second": 13.089,
-      "eval_steps_per_second": 1.643,
+      "eval_runtime": 36.5959,
+      "eval_samples_per_second": 13.499,
+      "eval_steps_per_second": 1.694,
       "step": 1400
     },
     {
@@ -333,10 +333,120 @@
     {
       "epoch": 2.493765586034913,
       "eval_loss": 0.2369847148656845,
-      "eval_runtime": 38.1397,
-      "eval_samples_per_second": 12.952,
-      "eval_steps_per_second": 1.626,
+      "eval_runtime": 36.5924,
+      "eval_samples_per_second": 13.5,
+      "eval_steps_per_second": 1.694,
       "step": 1500
+    },
+    {
+      "epoch": 2.57689110556941,
+      "grad_norm": 22271.5,
+      "learning_rate": 1.0657237936772047e-05,
+      "loss": 0.2684,
+      "step": 1550
+    },
+    {
+      "epoch": 2.660016625103907,
+      "grad_norm": 17982.9140625,
+      "learning_rate": 1.0033277870216307e-05,
+      "loss": 0.2753,
+      "step": 1600
+    },
+    {
+      "epoch": 2.660016625103907,
+      "eval_loss": 0.23503336310386658,
+      "eval_runtime": 36.5255,
+      "eval_samples_per_second": 13.525,
+      "eval_steps_per_second": 1.697,
+      "step": 1600
+    },
+    {
+      "epoch": 2.743142144638404,
+      "grad_norm": 20275.65625,
+      "learning_rate": 9.409317803660566e-06,
+      "loss": 0.282,
+      "step": 1650
+    },
+    {
+      "epoch": 2.826267664172901,
+      "grad_norm": 21899.2109375,
+      "learning_rate": 8.785357737104826e-06,
+      "loss": 0.2699,
+      "step": 1700
+    },
+    {
+      "epoch": 2.826267664172901,
+      "eval_loss": 0.23422521352767944,
+      "eval_runtime": 36.5068,
+      "eval_samples_per_second": 13.532,
+      "eval_steps_per_second": 1.698,
+      "step": 1700
+    },
+    {
+      "epoch": 2.9093931837073983,
+      "grad_norm": 17812.615234375,
+      "learning_rate": 8.161397670549084e-06,
+      "loss": 0.277,
+      "step": 1750
+    },
+    {
+      "epoch": 2.9925187032418954,
+      "grad_norm": 18110.896484375,
+      "learning_rate": 7.5374376039933445e-06,
+      "loss": 0.2673,
+      "step": 1800
+    },
+    {
+      "epoch": 2.9925187032418954,
+      "eval_loss": 0.23330263793468475,
+      "eval_runtime": 36.5181,
+      "eval_samples_per_second": 13.528,
+      "eval_steps_per_second": 1.698,
+      "step": 1800
+    },
+    {
+      "epoch": 3.0756442227763925,
+      "grad_norm": 64865.62109375,
+      "learning_rate": 6.913477537437604e-06,
+      "loss": 0.2742,
+      "step": 1850
+    },
+    {
+      "epoch": 3.1587697423108896,
+      "grad_norm": 22536.302734375,
+      "learning_rate": 6.289517470881864e-06,
+      "loss": 0.2723,
+      "step": 1900
+    },
+    {
+      "epoch": 3.1587697423108896,
+      "eval_loss": 0.23302872478961945,
+      "eval_runtime": 36.6159,
+      "eval_samples_per_second": 13.491,
+      "eval_steps_per_second": 1.693,
+      "step": 1900
+    },
+    {
+      "epoch": 3.2418952618453867,
+      "grad_norm": 26661.78125,
+      "learning_rate": 5.6655574043261234e-06,
+      "loss": 0.273,
+      "step": 1950
+    },
+    {
+      "epoch": 3.3250207813798838,
+      "grad_norm": 39719.59765625,
+      "learning_rate": 5.0415973377703825e-06,
+      "loss": 0.2746,
+      "step": 2000
+    },
+    {
+      "epoch": 3.3250207813798838,
+      "eval_loss": 0.23246362805366516,
+      "eval_runtime": 36.5215,
+      "eval_samples_per_second": 13.526,
+      "eval_steps_per_second": 1.698,
+      "step": 2000
     }
   ],
   "logging_steps": 50,
@@ -356,7 +466,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.461377145765888e+16,
+  "total_flos": 1.948482562424832e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null