diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..0c5d5cb83746ad04f33557861962126d86ff8913 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,13 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-385 filter=lfs diff=lfs merge=lfs -text
+config.json filter=lfs diff=lfs merge=lfs -text
+tokenizer_config.json filter=lfs diff=lfs merge=lfs -text
+model.safetensors filter=lfs diff=lfs merge=lfs -text
+special_tokens_map.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-100 filter=lfs diff=lfs merge=lfs -text
+checkpoint-200 filter=lfs diff=lfs merge=lfs -text
+checkpoint-300 filter=lfs diff=lfs merge=lfs -text
+generation_config.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoint-100/config.json b/checkpoint-100/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..4a3fd3ad7e0af9bc29478da576fa28f4642ce531
--- /dev/null
+++ b/checkpoint-100/config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322b2d3fc9de58487c02f1cd651d484d73244f9704de3d5feea36d002a38af0f
+size 748
diff --git a/checkpoint-100/generation_config.json b/checkpoint-100/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..475e140691313b73815f90f47ad5fde439467dfd
--- /dev/null
+++ b/checkpoint-100/generation_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82bdd024a0ba026e4001e0433d3922a27b1242913db7be87e2851d17165b99c8
+size 111
diff --git a/checkpoint-100/model.safetensors b/checkpoint-100/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..09167fd7228898dafafc108577c73a53ca80e2cb
--- /dev/null
+++ b/checkpoint-100/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c902b8779329b72d13123abd190eb8abc4874034542d6460c812f4c0c4ef2630
+size 810702192
diff --git a/checkpoint-100/optimizer.pt b/checkpoint-100/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b19701de06b234aef946dd03c21e8940224def9d
--- /dev/null
+++ b/checkpoint-100/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3aeba7985871af574775ffc9bc3f54f8a5465bd3fe06cabad919da69c90c5ad
+size 1621580026
diff --git a/checkpoint-100/rng_state.pth b/checkpoint-100/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4f427f074dcc36845a8b146ca04d57ca91cc33c9
--- /dev/null
+++ b/checkpoint-100/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c26b481b9cb8c27b6ce66e286d7b2720e0aefc60735ac6e7640899dd50d4bf55
+size 14244
diff --git a/checkpoint-100/scheduler.pt b/checkpoint-100/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3133bfee7b89e32cd717c406e196e6a4fb26212a
--- /dev/null
+++ b/checkpoint-100/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88d741f650ae118917922caa09145405cb8755dac6ba17ed9172d579629982fe
+size 1064
diff --git a/checkpoint-100/special_tokens_map.json b/checkpoint-100/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..2a565ac6033dc9066dd29eadd1019f36f9709c86
--- /dev/null
+++ b/checkpoint-100/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57491904f8680d4b52ed440f1f7ba48cad1c31ecf3eb453b03484e6ff4723ae8
+size 473
diff --git a/checkpoint-100/tokenizer.json b/checkpoint-100/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..614d0562ad8d2db17f5aa36ec866a872f7d821a0
--- /dev/null
+++ b/checkpoint-100/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:870f4e2baa6b683221fa52004d5d6f40ab8c9d31961617304b78c910c2c3caf2
+size 2114042
diff --git a/checkpoint-100/tokenizer_config.json b/checkpoint-100/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbc1d807e6afd67bc165abeb593b1c257efad4b3
--- /dev/null
+++ b/checkpoint-100/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4f587ac3a67c3964d3e84248f95ffc8e4f0e64d9de677692df39aa35eb2d11
+size 4845
diff --git a/checkpoint-100/trainer_state.json b/checkpoint-100/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..fbc2715e9c1699e6001f1f4e339b284275fe3211
--- /dev/null
+++ b/checkpoint-100/trainer_state.json
@@ -0,0 +1,823 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.25915127955944284,
+  "eval_steps": 10,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025915127955944283,
+      "grad_norm": 12.625,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 1.7977,
+      "step": 1
+    },
+    {
+      "epoch": 0.0051830255911888565,
+      "grad_norm": 12.0625,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 1.7787,
+      "step": 2
+    },
+    {
+      "epoch": 0.007774538386783284,
+      "grad_norm": 13.0,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 1.8003,
+      "step": 3
+    },
+    {
+      "epoch": 0.010366051182377713,
+      "grad_norm": 12.5625,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 1.787,
+      "step": 4
+    },
+    {
+      "epoch": 0.01295756397797214,
+      "grad_norm": 12.25,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.7684,
+      "step": 5
+    },
+    {
+      "epoch": 0.015549076773566569,
+      "grad_norm": 12.625,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 1.7544,
+      "step": 6
+    },
+    {
+      "epoch": 0.018140589569160998,
+      "grad_norm": 12.5,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 1.7769,
+      "step": 7
+    },
+    {
+      "epoch": 0.020732102364755426,
+      "grad_norm": 11.4375,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 1.7161,
+      "step": 8
+    },
+    {
+      "epoch": 0.023323615160349854,
+      "grad_norm": 11.5,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 1.7165,
+      "step": 9
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "grad_norm": 12.1875,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.7437,
+      "step": 10
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "eval_loss": 1.7256978750228882,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5814,
+      "eval_samples_per_second": 105.893,
+      "eval_steps_per_second": 13.262,
+      "step": 10
+    },
+    {
+      "epoch": 0.02850664075153871,
+      "grad_norm": 11.4375,
+      "learning_rate": 4.4e-06,
+      "loss": 1.7068,
+      "step": 11
+    },
+    {
+      "epoch": 0.031098153547133137,
+      "grad_norm": 11.75,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 1.688,
+      "step": 12
+    },
+    {
+      "epoch": 0.03368966634272757,
+      "grad_norm": 11.6875,
+      "learning_rate": 5.2e-06,
+      "loss": 1.6602,
+      "step": 13
+    },
+    {
+      "epoch": 0.036281179138321996,
+      "grad_norm": 11.5625,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 1.6389,
+      "step": 14
+    },
+    {
+      "epoch": 0.038872691933916424,
+      "grad_norm": 10.75,
+      "learning_rate": 6e-06,
+      "loss": 1.6228,
+      "step": 15
+    },
+    {
+      "epoch": 0.04146420472951085,
+      "grad_norm": 10.8125,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 1.5823,
+      "step": 16
+    },
+    {
+      "epoch": 0.04405571752510528,
+      "grad_norm": 9.5625,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 1.5134,
+      "step": 17
+    },
+    {
+      "epoch": 0.04664723032069971,
+      "grad_norm": 9.125,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 1.4833,
+      "step": 18
+    },
+    {
+      "epoch": 0.049238743116294136,
+      "grad_norm": 7.375,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 1.4966,
+      "step": 19
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "grad_norm": 7.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.4653,
+      "step": 20
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "eval_loss": 1.4041186571121216,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6226,
+      "eval_samples_per_second": 105.716,
+      "eval_steps_per_second": 13.24,
+      "step": 20
+    },
+    {
+      "epoch": 0.05442176870748299,
+      "grad_norm": 6.0625,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 1.3844,
+      "step": 21
+    },
+    {
+      "epoch": 0.05701328150307742,
+      "grad_norm": 5.375,
+      "learning_rate": 8.8e-06,
+      "loss": 1.3735,
+      "step": 22
+    },
+    {
+      "epoch": 0.05960479429867185,
+      "grad_norm": 5.09375,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 1.3281,
+      "step": 23
+    },
+    {
+      "epoch": 0.062196307094266275,
+      "grad_norm": 5.25,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 1.2837,
+      "step": 24
+    },
+    {
+      "epoch": 0.06478781988986071,
+      "grad_norm": 5.25,
+      "learning_rate": 1e-05,
+      "loss": 1.2777,
+      "step": 25
+    },
+    {
+      "epoch": 0.06737933268545514,
+      "grad_norm": 5.03125,
+      "learning_rate": 1.04e-05,
+      "loss": 1.2465,
+      "step": 26
+    },
+    {
+      "epoch": 0.06997084548104957,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.0800000000000002e-05,
+      "loss": 1.2041,
+      "step": 27
+    },
+    {
+      "epoch": 0.07256235827664399,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.1200000000000001e-05,
+      "loss": 1.1524,
+      "step": 28
+    },
+    {
+      "epoch": 0.07515387107223842,
+      "grad_norm": 4.65625,
+      "learning_rate": 1.16e-05,
+      "loss": 1.1491,
+      "step": 29
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "grad_norm": 5.125,
+      "learning_rate": 1.2e-05,
+      "loss": 1.127,
+      "step": 30
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "eval_loss": 1.105301856994629,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5195,
+      "eval_samples_per_second": 106.161,
+      "eval_steps_per_second": 13.296,
+      "step": 30
+    },
+    {
+      "epoch": 0.08033689666342728,
+      "grad_norm": 5.09375,
+      "learning_rate": 1.2400000000000002e-05,
+      "loss": 1.0993,
+      "step": 31
+    },
+    {
+      "epoch": 0.0829284094590217,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.2800000000000001e-05,
+      "loss": 1.0637,
+      "step": 32
+    },
+    {
+      "epoch": 0.08551992225461613,
+      "grad_norm": 4.75,
+      "learning_rate": 1.3200000000000002e-05,
+      "loss": 1.0447,
+      "step": 33
+    },
+    {
+      "epoch": 0.08811143505021056,
+      "grad_norm": 5.21875,
+      "learning_rate": 1.3600000000000002e-05,
+      "loss": 1.047,
+      "step": 34
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.4e-05,
+      "loss": 0.9841,
+      "step": 35
+    },
+    {
+      "epoch": 0.09329446064139942,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.4400000000000001e-05,
+      "loss": 0.9735,
+      "step": 36
+    },
+    {
+      "epoch": 0.09588597343699384,
+      "grad_norm": 5.71875,
+      "learning_rate": 1.48e-05,
+      "loss": 0.9409,
+      "step": 37
+    },
+    {
+      "epoch": 0.09847748623258827,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.5200000000000002e-05,
+      "loss": 0.9108,
+      "step": 38
+    },
+    {
+      "epoch": 0.1010689990281827,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.5600000000000003e-05,
+      "loss": 0.9029,
+      "step": 39
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.8938,
+      "step": 40
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "eval_loss": 0.8692858815193176,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5016,
+      "eval_samples_per_second": 106.238,
+      "eval_steps_per_second": 13.305,
+      "step": 40
+    },
+    {
+      "epoch": 0.10625202461937155,
+      "grad_norm": 5.125,
+      "learning_rate": 1.64e-05,
+      "loss": 0.8501,
+      "step": 41
+    },
+    {
+      "epoch": 0.10884353741496598,
+      "grad_norm": 4.53125,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 0.8545,
+      "step": 42
+    },
+    {
+      "epoch": 0.11143505021056041,
+      "grad_norm": 5.1875,
+      "learning_rate": 1.72e-05,
+      "loss": 0.8169,
+      "step": 43
+    },
+    {
+      "epoch": 0.11402656300615484,
+      "grad_norm": 4.5,
+      "learning_rate": 1.76e-05,
+      "loss": 0.8102,
+      "step": 44
+    },
+    {
+      "epoch": 0.11661807580174927,
+      "grad_norm": 4.3125,
+      "learning_rate": 1.8e-05,
+      "loss": 0.8093,
+      "step": 45
+    },
+    {
+      "epoch": 0.1192095885973437,
+      "grad_norm": 4.28125,
+      "learning_rate": 1.8400000000000003e-05,
+      "loss": 0.7798,
+      "step": 46
+    },
+    {
+      "epoch": 0.12180110139293812,
+      "grad_norm": 4.09375,
+      "learning_rate": 1.88e-05,
+      "loss": 0.7761,
+      "step": 47
+    },
+    {
+      "epoch": 0.12439261418853255,
+      "grad_norm": 4.125,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.7647,
+      "step": 48
+    },
+    {
+      "epoch": 0.12698412698412698,
+      "grad_norm": 3.765625,
+      "learning_rate": 1.9600000000000002e-05,
+      "loss": 0.7612,
+      "step": 49
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "grad_norm": 4.0,
+      "learning_rate": 2e-05,
+      "loss": 0.7498,
+      "step": 50
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "eval_loss": 0.744697093963623,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6181,
+      "eval_samples_per_second": 105.735,
+      "eval_steps_per_second": 13.242,
+      "step": 50
+    },
+    {
+      "epoch": 0.13216715257531583,
+      "grad_norm": 4.5,
+      "learning_rate": 1.9999560279257315e-05,
+      "loss": 0.7504,
+      "step": 51
+    },
+    {
+      "epoch": 0.13475866537091027,
+      "grad_norm": 3.796875,
+      "learning_rate": 1.999824115570012e-05,
+      "loss": 0.7181,
+      "step": 52
+    },
+    {
+      "epoch": 0.1373501781665047,
+      "grad_norm": 3.703125,
+      "learning_rate": 1.9996042745337617e-05,
+      "loss": 0.7195,
+      "step": 53
+    },
+    {
+      "epoch": 0.13994169096209913,
+      "grad_norm": 3.921875,
+      "learning_rate": 1.999296524150713e-05,
+      "loss": 0.7232,
+      "step": 54
+    },
+    {
+      "epoch": 0.14253320375769354,
+      "grad_norm": 3.6875,
+      "learning_rate": 1.9989008914857115e-05,
+      "loss": 0.725,
+      "step": 55
+    },
+    {
+      "epoch": 0.14512471655328799,
+      "grad_norm": 3.734375,
+      "learning_rate": 1.9984174113323352e-05,
+      "loss": 0.6972,
+      "step": 56
+    },
+    {
+      "epoch": 0.1477162293488824,
+      "grad_norm": 3.5,
+      "learning_rate": 1.9978461262098342e-05,
+      "loss": 0.6792,
+      "step": 57
+    },
+    {
+      "epoch": 0.15030774214447684,
+      "grad_norm": 4.0625,
+      "learning_rate": 1.9971870863593924e-05,
+      "loss": 0.7192,
+      "step": 58
+    },
+    {
+      "epoch": 0.15289925494007126,
+      "grad_norm": 3.75,
+      "learning_rate": 1.9964403497397086e-05,
+      "loss": 0.7107,
+      "step": 59
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9956059820218982e-05,
+      "loss": 0.6866,
+      "step": 60
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "eval_loss": 0.6849498152732849,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5203,
+      "eval_samples_per_second": 106.157,
+      "eval_steps_per_second": 13.295,
+      "step": 60
+    },
+    {
+      "epoch": 0.1580822805312601,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9946840565837206e-05,
+      "loss": 0.6818,
+      "step": 61
+    },
+    {
+      "epoch": 0.16067379332685455,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.9936746545031225e-05,
+      "loss": 0.6855,
+      "step": 62
+    },
+    {
+      "epoch": 0.16326530612244897,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9925778645511112e-05,
+      "loss": 0.6799,
+      "step": 63
+    },
+    {
+      "epoch": 0.1658568189180434,
+      "grad_norm": 3.828125,
+      "learning_rate": 1.991393783183945e-05,
+      "loss": 0.6719,
+      "step": 64
+    },
+    {
+      "epoch": 0.16844833171363785,
+      "grad_norm": 3.0,
+      "learning_rate": 1.990122514534651e-05,
+      "loss": 0.66,
+      "step": 65
+    },
+    {
+      "epoch": 0.17103984450923226,
+      "grad_norm": 2.96875,
+      "learning_rate": 1.988764170403869e-05,
+      "loss": 0.6718,
+      "step": 66
+    },
+    {
+      "epoch": 0.1736313573048267,
+      "grad_norm": 3.15625,
+      "learning_rate": 1.9873188702500163e-05,
+      "loss": 0.6661,
+      "step": 67
+    },
+    {
+      "epoch": 0.17622287010042112,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.9857867411787847e-05,
+      "loss": 0.6523,
+      "step": 68
+    },
+    {
+      "epoch": 0.17881438289601556,
+      "grad_norm": 3.203125,
+      "learning_rate": 1.9841679179319605e-05,
+      "loss": 0.6494,
+      "step": 69
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 2.875,
+      "learning_rate": 1.982462542875576e-05,
+      "loss": 0.6411,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "eval_loss": 0.6511064171791077,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5304,
+      "eval_samples_per_second": 106.113,
+      "eval_steps_per_second": 13.29,
+      "step": 70
+    },
+    {
+      "epoch": 0.18399740848720442,
+      "grad_norm": 3.296875,
+      "learning_rate": 1.9806707659873886e-05,
+      "loss": 0.6463,
+      "step": 71
+    },
+    {
+      "epoch": 0.18658892128279883,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9787927448436913e-05,
+      "loss": 0.6375,
+      "step": 72
+    },
+    {
+      "epoch": 0.18918043407839327,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9768286446054533e-05,
+      "loss": 0.6386,
+      "step": 73
+    },
+    {
+      "epoch": 0.1917719468739877,
+      "grad_norm": 2.875,
+      "learning_rate": 1.974778638003799e-05,
+      "loss": 0.6368,
+      "step": 74
+    },
+    {
+      "epoch": 0.19436345966958213,
+      "grad_norm": 2.953125,
+      "learning_rate": 1.972642905324813e-05,
+      "loss": 0.6271,
+      "step": 75
+    },
+    {
+      "epoch": 0.19695497246517654,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9704216343936875e-05,
+      "loss": 0.6501,
+      "step": 76
+    },
+    {
+      "epoch": 0.19954648526077098,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9681150205582025e-05,
+      "loss": 0.6111,
+      "step": 77
+    },
+    {
+      "epoch": 0.2021379980563654,
+      "grad_norm": 3.171875,
+      "learning_rate": 1.9657232666715485e-05,
+      "loss": 0.651,
+      "step": 78
+    },
+    {
+      "epoch": 0.20472951085195984,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9632465830744846e-05,
+      "loss": 0.6184,
+      "step": 79
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "grad_norm": 3.25,
+      "learning_rate": 1.9606851875768404e-05,
+      "loss": 0.6259,
+      "step": 80
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "eval_loss": 0.6278263330459595,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4892,
+      "eval_samples_per_second": 106.292,
+      "eval_steps_per_second": 13.312,
+      "step": 80
+    },
+    {
+      "epoch": 0.2099125364431487,
+      "grad_norm": 2.875,
+      "learning_rate": 1.9580393054383622e-05,
+      "loss": 0.6288,
+      "step": 81
+    },
+    {
+      "epoch": 0.2125040492387431,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9553091693489018e-05,
+      "loss": 0.6403,
+      "step": 82
+    },
+    {
+      "epoch": 0.21509556203433755,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9524950194079534e-05,
+      "loss": 0.621,
+      "step": 83
+    },
+    {
+      "epoch": 0.21768707482993196,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9495971031035367e-05,
+      "loss": 0.6135,
+      "step": 84
+    },
+    {
+      "epoch": 0.2202785876255264,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9466156752904344e-05,
+      "loss": 0.6211,
+      "step": 85
+    },
+    {
+      "epoch": 0.22287010042112082,
+      "grad_norm": 3.109375,
+      "learning_rate": 1.9435509981677764e-05,
+      "loss": 0.6198,
+      "step": 86
+    },
+    {
+      "epoch": 0.22546161321671526,
+      "grad_norm": 2.625,
+      "learning_rate": 1.9404033412559828e-05,
+      "loss": 0.5972,
+      "step": 87
+    },
+    {
+      "epoch": 0.22805312601230968,
+      "grad_norm": 3.0,
+      "learning_rate": 1.9371729813730606e-05,
+      "loss": 0.614,
+      "step": 88
+    },
+    {
+      "epoch": 0.23064463880790412,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9338602026102594e-05,
+      "loss": 0.5998,
+      "step": 89
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.9304652963070868e-05,
+      "loss": 0.6181,
+      "step": 90
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "eval_loss": 0.6108965277671814,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4982,
+      "eval_samples_per_second": 106.253,
+      "eval_steps_per_second": 13.307,
+      "step": 90
+    },
+    {
+      "epoch": 0.23582766439909297,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.9269885610256866e-05,
+      "loss": 0.6059,
+      "step": 91
+    },
+    {
+      "epoch": 0.2384191771946874,
+      "grad_norm": 2.734375,
+      "learning_rate": 1.9234303025245835e-05,
+      "loss": 0.6165,
+      "step": 92
+    },
+    {
+      "epoch": 0.24101068999028183,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.919790833731791e-05,
+      "loss": 0.6241,
+      "step": 93
+    },
+    {
+      "epoch": 0.24360220278587624,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9160704747172935e-05,
+      "loss": 0.5913,
+      "step": 94
+    },
+    {
+      "epoch": 0.24619371558147068,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9122695526648968e-05,
+      "loss": 0.585,
+      "step": 95
+    },
+    {
+      "epoch": 0.2487852283770651,
+      "grad_norm": 3.484375,
+      "learning_rate": 1.9083884018434545e-05,
+      "loss": 0.5897,
+      "step": 96
+    },
+    {
+      "epoch": 0.2513767411726595,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9044273635774707e-05,
+      "loss": 0.6032,
+      "step": 97
+    },
+    {
+      "epoch": 0.25396825396825395,
+      "grad_norm": 2.75,
+      "learning_rate": 1.900386786217083e-05,
+      "loss": 0.594,
+      "step": 98
+    },
+    {
+      "epoch": 0.2565597667638484,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.8962670251074276e-05,
+      "loss": 0.5958,
+      "step": 99
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.8920684425573865e-05,
+      "loss": 0.5821,
+      "step": 100
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "eval_loss": 0.5988509058952332,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4884,
+      "eval_samples_per_second": 106.295,
+      "eval_steps_per_second": 13.312,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 385,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.7825750933504e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-100/training_args.bin b/checkpoint-100/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f7aa13c9cfd007e1e40a3fab731420d1b2409529
--- /dev/null
+++ b/checkpoint-100/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca15ebb16a59dc055f8ae01b1eb1038e869a6556aa3b7b802c58f2dde6a2993c
+size 5304
diff --git a/checkpoint-200/config.json b/checkpoint-200/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..4a3fd3ad7e0af9bc29478da576fa28f4642ce531
--- /dev/null
+++ b/checkpoint-200/config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322b2d3fc9de58487c02f1cd651d484d73244f9704de3d5feea36d002a38af0f
+size 748
diff --git a/checkpoint-200/generation_config.json b/checkpoint-200/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..475e140691313b73815f90f47ad5fde439467dfd
--- /dev/null
+++ b/checkpoint-200/generation_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82bdd024a0ba026e4001e0433d3922a27b1242913db7be87e2851d17165b99c8
+size 111
diff --git a/checkpoint-200/model.safetensors b/checkpoint-200/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..29837505f95e0c72620fa15dfcd858d8f89cdb1c
--- /dev/null
+++ b/checkpoint-200/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7097f484aaede8efd96df5361bf11991048535719a9ece45a6160bf49c0121f9
+size 810702192
diff --git a/checkpoint-200/optimizer.pt b/checkpoint-200/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b04febd0aae3445b5947ede5c2b7c3378dcf1167
--- /dev/null
+++ b/checkpoint-200/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1891d4f0cfe566e718551469717420b08c3f586a4a837b4d47c08a779b670fc
+size 1621580026
diff --git a/checkpoint-200/rng_state.pth b/checkpoint-200/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ea17f6c5f674e0024fa4c4dc7b8e9d3161488eb2
--- /dev/null
+++ b/checkpoint-200/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aad5da87704f9496a4cad943d2aa8988a987b1e77b57fa4a28ff4c89ee480ac0
+size 14244
diff --git a/checkpoint-200/scheduler.pt b/checkpoint-200/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7a5eac6c52b04e3aaec52d56d623974e31959071
--- /dev/null
+++ b/checkpoint-200/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5efd9b5f266aeac292c1c263af3ec67abbe75dee3425023232f2c488a9f2e17e
+size 1064
diff --git a/checkpoint-200/special_tokens_map.json b/checkpoint-200/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..2a565ac6033dc9066dd29eadd1019f36f9709c86
--- /dev/null
+++ b/checkpoint-200/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57491904f8680d4b52ed440f1f7ba48cad1c31ecf3eb453b03484e6ff4723ae8
+size 473
diff --git a/checkpoint-200/tokenizer.json b/checkpoint-200/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..614d0562ad8d2db17f5aa36ec866a872f7d821a0
--- /dev/null
+++ b/checkpoint-200/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:870f4e2baa6b683221fa52004d5d6f40ab8c9d31961617304b78c910c2c3caf2
+size 2114042
diff --git a/checkpoint-200/tokenizer_config.json b/checkpoint-200/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbc1d807e6afd67bc165abeb593b1c257efad4b3
--- /dev/null
+++ b/checkpoint-200/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4f587ac3a67c3964d3e84248f95ffc8e4f0e64d9de677692df39aa35eb2d11
+size 4845
diff --git a/checkpoint-200/trainer_state.json b/checkpoint-200/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1477f4321c0c2a5dad2b0e6b77ffa21e98575540
--- /dev/null
+++ b/checkpoint-200/trainer_state.json
@@ -0,0 +1,1613 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.5183025591188857,
+  "eval_steps": 10,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025915127955944283,
+      "grad_norm": 12.625,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 1.7977,
+      "step": 1
+    },
+    {
+      "epoch": 0.0051830255911888565,
+      "grad_norm": 12.0625,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 1.7787,
+      "step": 2
+    },
+    {
+      "epoch": 0.007774538386783284,
+      "grad_norm": 13.0,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 1.8003,
+      "step": 3
+    },
+    {
+      "epoch": 0.010366051182377713,
+      "grad_norm": 12.5625,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 1.787,
+      "step": 4
+    },
+    {
+      "epoch": 0.01295756397797214,
+      "grad_norm": 12.25,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.7684,
+      "step": 5
+    },
+    {
+      "epoch": 0.015549076773566569,
+      "grad_norm": 12.625,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 1.7544,
+      "step": 6
+    },
+    {
+      "epoch": 0.018140589569160998,
+      "grad_norm": 12.5,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 1.7769,
+      "step": 7
+    },
+    {
+      "epoch": 0.020732102364755426,
+      "grad_norm": 11.4375,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 1.7161,
+      "step": 8
+    },
+    {
+      "epoch": 0.023323615160349854,
+      "grad_norm": 11.5,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 1.7165,
+      "step": 9
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "grad_norm": 12.1875,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.7437,
+      "step": 10
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "eval_loss": 1.7256978750228882,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5814,
+      "eval_samples_per_second": 105.893,
+      "eval_steps_per_second": 13.262,
+      "step": 10
+    },
+    {
+      "epoch": 0.02850664075153871,
+      "grad_norm": 11.4375,
+      "learning_rate": 4.4e-06,
+      "loss": 1.7068,
+      "step": 11
+    },
+    {
+      "epoch": 0.031098153547133137,
+      "grad_norm": 11.75,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 1.688,
+      "step": 12
+    },
+    {
+      "epoch": 0.03368966634272757,
+      "grad_norm": 11.6875,
+      "learning_rate": 5.2e-06,
+      "loss": 1.6602,
+      "step": 13
+    },
+    {
+      "epoch": 0.036281179138321996,
+      "grad_norm": 11.5625,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 1.6389,
+      "step": 14
+    },
+    {
+      "epoch": 0.038872691933916424,
+      "grad_norm": 10.75,
+      "learning_rate": 6e-06,
+      "loss": 1.6228,
+      "step": 15
+    },
+    {
+      "epoch": 0.04146420472951085,
+      "grad_norm": 10.8125,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 1.5823,
+      "step": 16
+    },
+    {
+      "epoch": 0.04405571752510528,
+      "grad_norm": 9.5625,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 1.5134,
+      "step": 17
+    },
+    {
+      "epoch": 0.04664723032069971,
+      "grad_norm": 9.125,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 1.4833,
+      "step": 18
+    },
+    {
+      "epoch": 0.049238743116294136,
+      "grad_norm": 7.375,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 1.4966,
+      "step": 19
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "grad_norm": 7.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.4653,
+      "step": 20
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "eval_loss": 1.4041186571121216,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6226,
+      "eval_samples_per_second": 105.716,
+      "eval_steps_per_second": 13.24,
+      "step": 20
+    },
+    {
+      "epoch": 0.05442176870748299,
+      "grad_norm": 6.0625,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 1.3844,
+      "step": 21
+    },
+    {
+      "epoch": 0.05701328150307742,
+      "grad_norm": 5.375,
+      "learning_rate": 8.8e-06,
+      "loss": 1.3735,
+      "step": 22
+    },
+    {
+      "epoch": 0.05960479429867185,
+      "grad_norm": 5.09375,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 1.3281,
+      "step": 23
+    },
+    {
+      "epoch": 0.062196307094266275,
+      "grad_norm": 5.25,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 1.2837,
+      "step": 24
+    },
+    {
+      "epoch": 0.06478781988986071,
+      "grad_norm": 5.25,
+      "learning_rate": 1e-05,
+      "loss": 1.2777,
+      "step": 25
+    },
+    {
+      "epoch": 0.06737933268545514,
+      "grad_norm": 5.03125,
+      "learning_rate": 1.04e-05,
+      "loss": 1.2465,
+      "step": 26
+    },
+    {
+      "epoch": 0.06997084548104957,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.0800000000000002e-05,
+      "loss": 1.2041,
+      "step": 27
+    },
+    {
+      "epoch": 0.07256235827664399,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.1200000000000001e-05,
+      "loss": 1.1524,
+      "step": 28
+    },
+    {
+      "epoch": 0.07515387107223842,
+      "grad_norm": 4.65625,
+      "learning_rate": 1.16e-05,
+      "loss": 1.1491,
+      "step": 29
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "grad_norm": 5.125,
+      "learning_rate": 1.2e-05,
+      "loss": 1.127,
+      "step": 30
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "eval_loss": 1.105301856994629,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5195,
+      "eval_samples_per_second": 106.161,
+      "eval_steps_per_second": 13.296,
+      "step": 30
+    },
+    {
+      "epoch": 0.08033689666342728,
+      "grad_norm": 5.09375,
+      "learning_rate": 1.2400000000000002e-05,
+      "loss": 1.0993,
+      "step": 31
+    },
+    {
+      "epoch": 0.0829284094590217,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.2800000000000001e-05,
+      "loss": 1.0637,
+      "step": 32
+    },
+    {
+      "epoch": 0.08551992225461613,
+      "grad_norm": 4.75,
+      "learning_rate": 1.3200000000000002e-05,
+      "loss": 1.0447,
+      "step": 33
+    },
+    {
+      "epoch": 0.08811143505021056,
+      "grad_norm": 5.21875,
+      "learning_rate": 1.3600000000000002e-05,
+      "loss": 1.047,
+      "step": 34
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.4e-05,
+      "loss": 0.9841,
+      "step": 35
+    },
+    {
+      "epoch": 0.09329446064139942,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.4400000000000001e-05,
+      "loss": 0.9735,
+      "step": 36
+    },
+    {
+      "epoch": 0.09588597343699384,
+      "grad_norm": 5.71875,
+      "learning_rate": 1.48e-05,
+      "loss": 0.9409,
+      "step": 37
+    },
+    {
+      "epoch": 0.09847748623258827,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.5200000000000002e-05,
+      "loss": 0.9108,
+      "step": 38
+    },
+    {
+      "epoch": 0.1010689990281827,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.5600000000000003e-05,
+      "loss": 0.9029,
+      "step": 39
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.8938,
+      "step": 40
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "eval_loss": 0.8692858815193176,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5016,
+      "eval_samples_per_second": 106.238,
+      "eval_steps_per_second": 13.305,
+      "step": 40
+    },
+    {
+      "epoch": 0.10625202461937155,
+      "grad_norm": 5.125,
+      "learning_rate": 1.64e-05,
+      "loss": 0.8501,
+      "step": 41
+    },
+    {
+      "epoch": 0.10884353741496598,
+      "grad_norm": 4.53125,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 0.8545,
+      "step": 42
+    },
+    {
+      "epoch": 0.11143505021056041,
+      "grad_norm": 5.1875,
+      "learning_rate": 1.72e-05,
+      "loss": 0.8169,
+      "step": 43
+    },
+    {
+      "epoch": 0.11402656300615484,
+      "grad_norm": 4.5,
+      "learning_rate": 1.76e-05,
+      "loss": 0.8102,
+      "step": 44
+    },
+    {
+      "epoch": 0.11661807580174927,
+      "grad_norm": 4.3125,
+      "learning_rate": 1.8e-05,
+      "loss": 0.8093,
+      "step": 45
+    },
+    {
+      "epoch": 0.1192095885973437,
+      "grad_norm": 4.28125,
+      "learning_rate": 1.8400000000000003e-05,
+      "loss": 0.7798,
+      "step": 46
+    },
+    {
+      "epoch": 0.12180110139293812,
+      "grad_norm": 4.09375,
+      "learning_rate": 1.88e-05,
+      "loss": 0.7761,
+      "step": 47
+    },
+    {
+      "epoch": 0.12439261418853255,
+      "grad_norm": 4.125,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.7647,
+      "step": 48
+    },
+    {
+      "epoch": 0.12698412698412698,
+      "grad_norm": 3.765625,
+      "learning_rate": 1.9600000000000002e-05,
+      "loss": 0.7612,
+      "step": 49
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "grad_norm": 4.0,
+      "learning_rate": 2e-05,
+      "loss": 0.7498,
+      "step": 50
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "eval_loss": 0.744697093963623,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6181,
+      "eval_samples_per_second": 105.735,
+      "eval_steps_per_second": 13.242,
+      "step": 50
+    },
+    {
+      "epoch": 0.13216715257531583,
+      "grad_norm": 4.5,
+      "learning_rate": 1.9999560279257315e-05,
+      "loss": 0.7504,
+      "step": 51
+    },
+    {
+      "epoch": 0.13475866537091027,
+      "grad_norm": 3.796875,
+      "learning_rate": 1.999824115570012e-05,
+      "loss": 0.7181,
+      "step": 52
+    },
+    {
+      "epoch": 0.1373501781665047,
+      "grad_norm": 3.703125,
+      "learning_rate": 1.9996042745337617e-05,
+      "loss": 0.7195,
+      "step": 53
+    },
+    {
+      "epoch": 0.13994169096209913,
+      "grad_norm": 3.921875,
+      "learning_rate": 1.999296524150713e-05,
+      "loss": 0.7232,
+      "step": 54
+    },
+    {
+      "epoch": 0.14253320375769354,
+      "grad_norm": 3.6875,
+      "learning_rate": 1.9989008914857115e-05,
+      "loss": 0.725,
+      "step": 55
+    },
+    {
+      "epoch": 0.14512471655328799,
+      "grad_norm": 3.734375,
+      "learning_rate": 1.9984174113323352e-05,
+      "loss": 0.6972,
+      "step": 56
+    },
+    {
+      "epoch": 0.1477162293488824,
+      "grad_norm": 3.5,
+      "learning_rate": 1.9978461262098342e-05,
+      "loss": 0.6792,
+      "step": 57
+    },
+    {
+      "epoch": 0.15030774214447684,
+      "grad_norm": 4.0625,
+      "learning_rate": 1.9971870863593924e-05,
+      "loss": 0.7192,
+      "step": 58
+    },
+    {
+      "epoch": 0.15289925494007126,
+      "grad_norm": 3.75,
+      "learning_rate": 1.9964403497397086e-05,
+      "loss": 0.7107,
+      "step": 59
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9956059820218982e-05,
+      "loss": 0.6866,
+      "step": 60
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "eval_loss": 0.6849498152732849,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5203,
+      "eval_samples_per_second": 106.157,
+      "eval_steps_per_second": 13.295,
+      "step": 60
+    },
+    {
+      "epoch": 0.1580822805312601,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9946840565837206e-05,
+      "loss": 0.6818,
+      "step": 61
+    },
+    {
+      "epoch": 0.16067379332685455,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.9936746545031225e-05,
+      "loss": 0.6855,
+      "step": 62
+    },
+    {
+      "epoch": 0.16326530612244897,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9925778645511112e-05,
+      "loss": 0.6799,
+      "step": 63
+    },
+    {
+      "epoch": 0.1658568189180434,
+      "grad_norm": 3.828125,
+      "learning_rate": 1.991393783183945e-05,
+      "loss": 0.6719,
+      "step": 64
+    },
+    {
+      "epoch": 0.16844833171363785,
+      "grad_norm": 3.0,
+      "learning_rate": 1.990122514534651e-05,
+      "loss": 0.66,
+      "step": 65
+    },
+    {
+      "epoch": 0.17103984450923226,
+      "grad_norm": 2.96875,
+      "learning_rate": 1.988764170403869e-05,
+      "loss": 0.6718,
+      "step": 66
+    },
+    {
+      "epoch": 0.1736313573048267,
+      "grad_norm": 3.15625,
+      "learning_rate": 1.9873188702500163e-05,
+      "loss": 0.6661,
+      "step": 67
+    },
+    {
+      "epoch": 0.17622287010042112,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.9857867411787847e-05,
+      "loss": 0.6523,
+      "step": 68
+    },
+    {
+      "epoch": 0.17881438289601556,
+      "grad_norm": 3.203125,
+      "learning_rate": 1.9841679179319605e-05,
+      "loss": 0.6494,
+      "step": 69
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 2.875,
+      "learning_rate": 1.982462542875576e-05,
+      "loss": 0.6411,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "eval_loss": 0.6511064171791077,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5304,
+      "eval_samples_per_second": 106.113,
+      "eval_steps_per_second": 13.29,
+      "step": 70
+    },
+    {
+      "epoch": 0.18399740848720442,
+      "grad_norm": 3.296875,
+      "learning_rate": 1.9806707659873886e-05,
+      "loss": 0.6463,
+      "step": 71
+    },
+    {
+      "epoch": 0.18658892128279883,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9787927448436913e-05,
+      "loss": 0.6375,
+      "step": 72
+    },
+    {
+      "epoch": 0.18918043407839327,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9768286446054533e-05,
+      "loss": 0.6386,
+      "step": 73
+    },
+    {
+      "epoch": 0.1917719468739877,
+      "grad_norm": 2.875,
+      "learning_rate": 1.974778638003799e-05,
+      "loss": 0.6368,
+      "step": 74
+    },
+    {
+      "epoch": 0.19436345966958213,
+      "grad_norm": 2.953125,
+      "learning_rate": 1.972642905324813e-05,
+      "loss": 0.6271,
+      "step": 75
+    },
+    {
+      "epoch": 0.19695497246517654,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9704216343936875e-05,
+      "loss": 0.6501,
+      "step": 76
+    },
+    {
+      "epoch": 0.19954648526077098,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9681150205582025e-05,
+      "loss": 0.6111,
+      "step": 77
+    },
+    {
+      "epoch": 0.2021379980563654,
+      "grad_norm": 3.171875,
+      "learning_rate": 1.9657232666715485e-05,
+      "loss": 0.651,
+      "step": 78
+    },
+    {
+      "epoch": 0.20472951085195984,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9632465830744846e-05,
+      "loss": 0.6184,
+      "step": 79
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "grad_norm": 3.25,
+      "learning_rate": 1.9606851875768404e-05,
+      "loss": 0.6259,
+      "step": 80
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "eval_loss": 0.6278263330459595,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4892,
+      "eval_samples_per_second": 106.292,
+      "eval_steps_per_second": 13.312,
+      "step": 80
+    },
+    {
+      "epoch": 0.2099125364431487,
+      "grad_norm": 2.875,
+      "learning_rate": 1.9580393054383622e-05,
+      "loss": 0.6288,
+      "step": 81
+    },
+    {
+      "epoch": 0.2125040492387431,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9553091693489018e-05,
+      "loss": 0.6403,
+      "step": 82
+    },
+    {
+      "epoch": 0.21509556203433755,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9524950194079534e-05,
+      "loss": 0.621,
+      "step": 83
+    },
+    {
+      "epoch": 0.21768707482993196,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9495971031035367e-05,
+      "loss": 0.6135,
+      "step": 84
+    },
+    {
+      "epoch": 0.2202785876255264,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9466156752904344e-05,
+      "loss": 0.6211,
+      "step": 85
+    },
+    {
+      "epoch": 0.22287010042112082,
+      "grad_norm": 3.109375,
+      "learning_rate": 1.9435509981677764e-05,
+      "loss": 0.6198,
+      "step": 86
+    },
+    {
+      "epoch": 0.22546161321671526,
+      "grad_norm": 2.625,
+      "learning_rate": 1.9404033412559828e-05,
+      "loss": 0.5972,
+      "step": 87
+    },
+    {
+      "epoch": 0.22805312601230968,
+      "grad_norm": 3.0,
+      "learning_rate": 1.9371729813730606e-05,
+      "loss": 0.614,
+      "step": 88
+    },
+    {
+      "epoch": 0.23064463880790412,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9338602026102594e-05,
+      "loss": 0.5998,
+      "step": 89
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.9304652963070868e-05,
+      "loss": 0.6181,
+      "step": 90
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "eval_loss": 0.6108965277671814,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4982,
+      "eval_samples_per_second": 106.253,
+      "eval_steps_per_second": 13.307,
+      "step": 90
+    },
+    {
+      "epoch": 0.23582766439909297,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.9269885610256866e-05,
+      "loss": 0.6059,
+      "step": 91
+    },
+    {
+      "epoch": 0.2384191771946874,
+      "grad_norm": 2.734375,
+      "learning_rate": 1.9234303025245835e-05,
+      "loss": 0.6165,
+      "step": 92
+    },
+    {
+      "epoch": 0.24101068999028183,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.919790833731791e-05,
+      "loss": 0.6241,
+      "step": 93
+    },
+    {
+      "epoch": 0.24360220278587624,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9160704747172935e-05,
+      "loss": 0.5913,
+      "step": 94
+    },
+    {
+      "epoch": 0.24619371558147068,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9122695526648968e-05,
+      "loss": 0.585,
+      "step": 95
+    },
+    {
+      "epoch": 0.2487852283770651,
+      "grad_norm": 3.484375,
+      "learning_rate": 1.9083884018434545e-05,
+      "loss": 0.5897,
+      "step": 96
+    },
+    {
+      "epoch": 0.2513767411726595,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9044273635774707e-05,
+      "loss": 0.6032,
+      "step": 97
+    },
+    {
+      "epoch": 0.25396825396825395,
+      "grad_norm": 2.75,
+      "learning_rate": 1.900386786217083e-05,
+      "loss": 0.594,
+      "step": 98
+    },
+    {
+      "epoch": 0.2565597667638484,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.8962670251074276e-05,
+      "loss": 0.5958,
+      "step": 99
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.8920684425573865e-05,
+      "loss": 0.5821,
+      "step": 100
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "eval_loss": 0.5988509058952332,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4884,
+      "eval_samples_per_second": 106.295,
+      "eval_steps_per_second": 13.312,
+      "step": 100
+    },
+    {
+      "epoch": 0.2617427923550373,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.887791407807728e-05,
+      "loss": 0.5904,
+      "step": 101
+    },
+    {
+      "epoch": 0.26433430515063167,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.883436296998631e-05,
+      "loss": 0.5913,
+      "step": 102
+    },
+    {
+      "epoch": 0.2669258179462261,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.8790034931366073e-05,
+      "loss": 0.6114,
+      "step": 103
+    },
+    {
+      "epoch": 0.26951733074182055,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.8744933860608183e-05,
+      "loss": 0.5803,
+      "step": 104
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 3.40625,
+      "learning_rate": 1.8699063724087905e-05,
+      "loss": 0.5865,
+      "step": 105
+    },
+    {
+      "epoch": 0.2747003563330094,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.865242855581534e-05,
+      "loss": 0.586,
+      "step": 106
+    },
+    {
+      "epoch": 0.2772918691286038,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.8605032457080655e-05,
+      "loss": 0.5926,
+      "step": 107
+    },
+    {
+      "epoch": 0.27988338192419826,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.8556879596093393e-05,
+      "loss": 0.5927,
+      "step": 108
+    },
+    {
+      "epoch": 0.2824748947197927,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.850797420761592e-05,
+      "loss": 0.5937,
+      "step": 109
+    },
+    {
+      "epoch": 0.2850664075153871,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.8458320592590976e-05,
+      "loss": 0.5644,
+      "step": 110
+    },
+    {
+      "epoch": 0.2850664075153871,
+      "eval_loss": 0.5878391265869141,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.539,
+      "eval_samples_per_second": 106.076,
+      "eval_steps_per_second": 13.285,
+      "step": 110
+    },
+    {
+      "epoch": 0.28765792031098153,
+      "grad_norm": 2.75,
+      "learning_rate": 1.840792311776346e-05,
+      "loss": 0.5783,
+      "step": 111
+    },
+    {
+      "epoch": 0.29024943310657597,
+      "grad_norm": 2.375,
+      "learning_rate": 1.8356786215296387e-05,
+      "loss": 0.5813,
+      "step": 112
+    },
+    {
+      "epoch": 0.2928409459021704,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.83049143823811e-05,
+      "loss": 0.5788,
+      "step": 113
+    },
+    {
+      "epoch": 0.2954324586977648,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.825231218084178e-05,
+      "loss": 0.5831,
+      "step": 114
+    },
+    {
+      "epoch": 0.29802397149335924,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.8198984236734246e-05,
+      "loss": 0.5715,
+      "step": 115
+    },
+    {
+      "epoch": 0.3006154842889537,
+      "grad_norm": 2.546875,
+      "learning_rate": 1.8144935239939147e-05,
+      "loss": 0.5938,
+      "step": 116
+    },
+    {
+      "epoch": 0.3032069970845481,
+      "grad_norm": 2.90625,
+      "learning_rate": 1.8090169943749477e-05,
+      "loss": 0.5911,
+      "step": 117
+    },
+    {
+      "epoch": 0.3057985098801425,
+      "grad_norm": 2.859375,
+      "learning_rate": 1.8034693164452578e-05,
+      "loss": 0.5815,
+      "step": 118
+    },
+    {
+      "epoch": 0.30839002267573695,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.7978509780906578e-05,
+      "loss": 0.5804,
+      "step": 119
+    },
+    {
+      "epoch": 0.3109815354713314,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.7921624734111292e-05,
+      "loss": 0.5749,
+      "step": 120
+    },
+    {
+      "epoch": 0.3109815354713314,
+      "eval_loss": 0.5794549584388733,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5897,
+      "eval_samples_per_second": 105.857,
+      "eval_steps_per_second": 13.258,
+      "step": 120
+    },
+    {
+      "epoch": 0.31357304826692584,
+      "grad_norm": 2.703125,
+      "learning_rate": 1.786404302677374e-05,
+      "loss": 0.5873,
+      "step": 121
+    },
+    {
+      "epoch": 0.3161645610625202,
+      "grad_norm": 2.375,
+      "learning_rate": 1.7805769722868132e-05,
+      "loss": 0.5856,
+      "step": 122
+    },
+    {
+      "epoch": 0.31875607385811466,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.7746809947190567e-05,
+      "loss": 0.578,
+      "step": 123
+    },
+    {
+      "epoch": 0.3213475866537091,
+      "grad_norm": 3.421875,
+      "learning_rate": 1.768716888490832e-05,
+      "loss": 0.5566,
+      "step": 124
+    },
+    {
+      "epoch": 0.32393909944930355,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.762685178110382e-05,
+      "loss": 0.5638,
+      "step": 125
+    },
+    {
+      "epoch": 0.32653061224489793,
+      "grad_norm": 3.125,
+      "learning_rate": 1.7565863940313416e-05,
+      "loss": 0.5827,
+      "step": 126
+    },
+    {
+      "epoch": 0.3291221250404924,
+      "grad_norm": 3.1875,
+      "learning_rate": 1.7504210726060828e-05,
+      "loss": 0.5596,
+      "step": 127
+    },
+    {
+      "epoch": 0.3317136378360868,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.7441897560385493e-05,
+      "loss": 0.5768,
+      "step": 128
+    },
+    {
+      "epoch": 0.33430515063168126,
+      "grad_norm": 2.546875,
+      "learning_rate": 1.7378929923365706e-05,
+      "loss": 0.5732,
+      "step": 129
+    },
+    {
+      "epoch": 0.3368966634272757,
+      "grad_norm": 2.78125,
+      "learning_rate": 1.731531335263669e-05,
+      "loss": 0.5781,
+      "step": 130
+    },
+    {
+      "epoch": 0.3368966634272757,
+      "eval_loss": 0.572753369808197,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5459,
+      "eval_samples_per_second": 106.046,
+      "eval_steps_per_second": 13.281,
+      "step": 130
+    },
+    {
+      "epoch": 0.3394881762228701,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.7251053442903597e-05,
+      "loss": 0.5689,
+      "step": 131
+    },
+    {
+      "epoch": 0.34207968901846453,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.7186155845449465e-05,
+      "loss": 0.5695,
+      "step": 132
+    },
+    {
+      "epoch": 0.34467120181405897,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.712062626763825e-05,
+      "loss": 0.5627,
+      "step": 133
+    },
+    {
+      "epoch": 0.3472627146096534,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.7054470472412876e-05,
+      "loss": 0.5759,
+      "step": 134
+    },
+    {
+      "epoch": 0.3498542274052478,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.698769427778842e-05,
+      "loss": 0.5662,
+      "step": 135
+    },
+    {
+      "epoch": 0.35244574020084224,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.6920303556340462e-05,
+      "loss": 0.5671,
+      "step": 136
+    },
+    {
+      "epoch": 0.3550372529964367,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.6852304234688627e-05,
+      "loss": 0.5572,
+      "step": 137
+    },
+    {
+      "epoch": 0.3576287657920311,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.678370229297535e-05,
+      "loss": 0.5796,
+      "step": 138
+    },
+    {
+      "epoch": 0.3602202785876255,
+      "grad_norm": 3.03125,
+      "learning_rate": 1.6714503764339986e-05,
+      "loss": 0.5642,
+      "step": 139
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.664471473438822e-05,
+      "loss": 0.5711,
+      "step": 140
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "eval_loss": 0.566669762134552,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5724,
+      "eval_samples_per_second": 105.932,
+      "eval_steps_per_second": 13.267,
+      "step": 140
+    },
+    {
+      "epoch": 0.3654033041788144,
+      "grad_norm": 2.625,
+      "learning_rate": 1.657434134065686e-05,
+      "loss": 0.5559,
+      "step": 141
+    },
+    {
+      "epoch": 0.36799481697440883,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.6503389772074104e-05,
+      "loss": 0.575,
+      "step": 142
+    },
+    {
+      "epoch": 0.3705863297700032,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.643186626841524e-05,
+      "loss": 0.5851,
+      "step": 143
+    },
+    {
+      "epoch": 0.37317784256559766,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.6359777119753887e-05,
+      "loss": 0.559,
+      "step": 144
+    },
+    {
+      "epoch": 0.3757693553611921,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.628712866590885e-05,
+      "loss": 0.5759,
+      "step": 145
+    },
+    {
+      "epoch": 0.37836086815678655,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.6213927295886547e-05,
+      "loss": 0.5579,
+      "step": 146
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.6140179447319134e-05,
+      "loss": 0.5717,
+      "step": 147
+    },
+    {
+      "epoch": 0.3835438937479754,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.6065891605898358e-05,
+      "loss": 0.5531,
+      "step": 148
+    },
+    {
+      "epoch": 0.3861354065435698,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.5991070304805184e-05,
+      "loss": 0.5358,
+      "step": 149
+    },
+    {
+      "epoch": 0.38872691933916426,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.5915722124135227e-05,
+      "loss": 0.5715,
+      "step": 150
+    },
+    {
+      "epoch": 0.38872691933916426,
+      "eval_loss": 0.5618191361427307,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5943,
+      "eval_samples_per_second": 105.838,
+      "eval_steps_per_second": 13.255,
+      "step": 150
+    },
+    {
+      "epoch": 0.39131843213475864,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.5839853690320074e-05,
+      "loss": 0.5413,
+      "step": 151
+    },
+    {
+      "epoch": 0.3939099449303531,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.5763471675544547e-05,
+      "loss": 0.5652,
+      "step": 152
+    },
+    {
+      "epoch": 0.3965014577259475,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.5686582797159892e-05,
+      "loss": 0.5682,
+      "step": 153
+    },
+    {
+      "epoch": 0.39909297052154197,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.5609193817093057e-05,
+      "loss": 0.5577,
+      "step": 154
+    },
+    {
+      "epoch": 0.40168448331713635,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.5531311541251995e-05,
+      "loss": 0.5412,
+      "step": 155
+    },
+    {
+      "epoch": 0.4042759961127308,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.5452942818927144e-05,
+      "loss": 0.5441,
+      "step": 156
+    },
+    {
+      "epoch": 0.40686750890832524,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.5374094542189055e-05,
+      "loss": 0.5764,
+      "step": 157
+    },
+    {
+      "epoch": 0.4094590217039197,
+      "grad_norm": 3.078125,
+      "learning_rate": 1.52947736452823e-05,
+      "loss": 0.5763,
+      "step": 158
+    },
+    {
+      "epoch": 0.41205053449951407,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.521498710401561e-05,
+      "loss": 0.5482,
+      "step": 159
+    },
+    {
+      "epoch": 0.4146420472951085,
+      "grad_norm": 2.375,
+      "learning_rate": 1.513474193514842e-05,
+      "loss": 0.559,
+      "step": 160
+    },
+    {
+      "epoch": 0.4146420472951085,
+      "eval_loss": 0.5573883056640625,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5874,
+      "eval_samples_per_second": 105.867,
+      "eval_steps_per_second": 13.259,
+      "step": 160
+    },
+    {
+      "epoch": 0.41723356009070295,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.505404519577379e-05,
+      "loss": 0.5642,
+      "step": 161
+    },
+    {
+      "epoch": 0.4198250728862974,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.4972903982697744e-05,
+      "loss": 0.5535,
+      "step": 162
+    },
+    {
+      "epoch": 0.42241658568189183,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.4891325431815182e-05,
+      "loss": 0.5519,
+      "step": 163
+    },
+    {
+      "epoch": 0.4250080984774862,
+      "grad_norm": 2.0,
+      "learning_rate": 1.48093167174823e-05,
+      "loss": 0.5611,
+      "step": 164
+    },
+    {
+      "epoch": 0.42759961127308066,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.4726885051885654e-05,
+      "loss": 0.5556,
+      "step": 165
+    },
+    {
+      "epoch": 0.4301911240686751,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.4644037684407883e-05,
+      "loss": 0.5289,
+      "step": 166
+    },
+    {
+      "epoch": 0.43278263686426954,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.4560781900990187e-05,
+      "loss": 0.5538,
+      "step": 167
+    },
+    {
+      "epoch": 0.43537414965986393,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.4477125023491536e-05,
+      "loss": 0.5589,
+      "step": 168
+    },
+    {
+      "epoch": 0.43796566245545837,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.4393074409044802e-05,
+      "loss": 0.5423,
+      "step": 169
+    },
+    {
+      "epoch": 0.4405571752510528,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.4308637449409705e-05,
+      "loss": 0.5593,
+      "step": 170
+    },
+    {
+      "epoch": 0.4405571752510528,
+      "eval_loss": 0.553532063961029,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5188,
+      "eval_samples_per_second": 106.163,
+      "eval_steps_per_second": 13.296,
+      "step": 170
+    },
+    {
+      "epoch": 0.44314868804664725,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.4223821570322762e-05,
+      "loss": 0.5789,
+      "step": 171
+    },
+    {
+      "epoch": 0.44574020084224164,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.4138634230844242e-05,
+      "loss": 0.5693,
+      "step": 172
+    },
+    {
+      "epoch": 0.4483317136378361,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.4053082922702183e-05,
+      "loss": 0.5496,
+      "step": 173
+    },
+    {
+      "epoch": 0.4509232264334305,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.3967175169633537e-05,
+      "loss": 0.5362,
+      "step": 174
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.3880918526722497e-05,
+      "loss": 0.5739,
+      "step": 175
+    },
+    {
+      "epoch": 0.45610625202461935,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.3794320579736084e-05,
+      "loss": 0.5493,
+      "step": 176
+    },
+    {
+      "epoch": 0.4586977648202138,
+      "grad_norm": 2.59375,
+      "learning_rate": 1.3707388944457007e-05,
+      "loss": 0.5665,
+      "step": 177
+    },
+    {
+      "epoch": 0.46128927761580824,
+      "grad_norm": 2.125,
+      "learning_rate": 1.3620131266013912e-05,
+      "loss": 0.5557,
+      "step": 178
+    },
+    {
+      "epoch": 0.4638807904114027,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.3532555218209037e-05,
+      "loss": 0.5402,
+      "step": 179
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "grad_norm": 2.375,
+      "learning_rate": 1.344466850284333e-05,
+      "loss": 0.5625,
+      "step": 180
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "eval_loss": 0.5502737164497375,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6233,
+      "eval_samples_per_second": 105.713,
+      "eval_steps_per_second": 13.239,
+      "step": 180
+    },
+    {
+      "epoch": 0.4690638160025915,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.335647884903915e-05,
+      "loss": 0.5492,
+      "step": 181
+    },
+    {
+      "epoch": 0.47165532879818595,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.3267994012560505e-05,
+      "loss": 0.5493,
+      "step": 182
+    },
+    {
+      "epoch": 0.4742468415937804,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.3179221775131005e-05,
+      "loss": 0.5508,
+      "step": 183
+    },
+    {
+      "epoch": 0.4768383543893748,
+      "grad_norm": 3.34375,
+      "learning_rate": 1.3090169943749475e-05,
+      "loss": 0.5427,
+      "step": 184
+    },
+    {
+      "epoch": 0.4794298671849692,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.300084635000341e-05,
+      "loss": 0.568,
+      "step": 185
+    },
+    {
+      "epoch": 0.48202137998056366,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.2911258849380199e-05,
+      "loss": 0.5478,
+      "step": 186
+    },
+    {
+      "epoch": 0.4846128927761581,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.2821415320576309e-05,
+      "loss": 0.5582,
+      "step": 187
+    },
+    {
+      "epoch": 0.4872044055717525,
+      "grad_norm": 3.0,
+      "learning_rate": 1.273132366480438e-05,
+      "loss": 0.5368,
+      "step": 188
+    },
+    {
+      "epoch": 0.4897959183673469,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.2640991805098369e-05,
+      "loss": 0.5608,
+      "step": 189
+    },
+    {
+      "epoch": 0.49238743116294137,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.2550427685616767e-05,
+      "loss": 0.5482,
+      "step": 190
+    },
+    {
+      "epoch": 0.49238743116294137,
+      "eval_loss": 0.547302782535553,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5429,
+      "eval_samples_per_second": 106.059,
+      "eval_steps_per_second": 13.283,
+      "step": 190
+    },
+    {
+      "epoch": 0.4949789439585358,
+      "grad_norm": 2.640625,
+      "learning_rate": 1.2459639270943944e-05,
+      "loss": 0.5508,
+      "step": 191
+    },
+    {
+      "epoch": 0.4975704567541302,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.2368634545389732e-05,
+      "loss": 0.5402,
+      "step": 192
+    },
+    {
+      "epoch": 0.5001619695497247,
+      "grad_norm": 2.25,
+      "learning_rate": 1.2277421512287227e-05,
+      "loss": 0.5313,
+      "step": 193
+    },
+    {
+      "epoch": 0.502753482345319,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.2186008193288963e-05,
+      "loss": 0.5535,
+      "step": 194
+    },
+    {
+      "epoch": 0.5053449951409135,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.2094402627661447e-05,
+      "loss": 0.5375,
+      "step": 195
+    },
+    {
+      "epoch": 0.5079365079365079,
+      "grad_norm": 2.375,
+      "learning_rate": 1.2002612871578143e-05,
+      "loss": 0.5564,
+      "step": 196
+    },
+    {
+      "epoch": 0.5105280207321024,
+      "grad_norm": 2.5,
+      "learning_rate": 1.1910646997411002e-05,
+      "loss": 0.5459,
+      "step": 197
+    },
+    {
+      "epoch": 0.5131195335276968,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.1818513093020514e-05,
+      "loss": 0.5545,
+      "step": 198
+    },
+    {
+      "epoch": 0.5157110463232912,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.172621926104446e-05,
+      "loss": 0.535,
+      "step": 199
+    },
+    {
+      "epoch": 0.5183025591188857,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.1633773618185302e-05,
+      "loss": 0.5469,
+      "step": 200
+    },
+    {
+      "epoch": 0.5183025591188857,
+      "eval_loss": 0.545219898223877,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4828,
+      "eval_samples_per_second": 106.319,
+      "eval_steps_per_second": 13.315,
+      "step": 200
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 385,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.5651501867008e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-200/training_args.bin b/checkpoint-200/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f7aa13c9cfd007e1e40a3fab731420d1b2409529
--- /dev/null
+++ b/checkpoint-200/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca15ebb16a59dc055f8ae01b1eb1038e869a6556aa3b7b802c58f2dde6a2993c
+size 5304
diff --git a/checkpoint-300/config.json b/checkpoint-300/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..4a3fd3ad7e0af9bc29478da576fa28f4642ce531
--- /dev/null
+++ b/checkpoint-300/config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322b2d3fc9de58487c02f1cd651d484d73244f9704de3d5feea36d002a38af0f
+size 748
diff --git a/checkpoint-300/generation_config.json b/checkpoint-300/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..475e140691313b73815f90f47ad5fde439467dfd
--- /dev/null
+++ b/checkpoint-300/generation_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82bdd024a0ba026e4001e0433d3922a27b1242913db7be87e2851d17165b99c8
+size 111
diff --git a/checkpoint-300/model.safetensors b/checkpoint-300/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a07e9b419c8d48cf5d818d3137858d2cb7ce9ea7
--- /dev/null
+++ b/checkpoint-300/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0aae1e28989d1966e9e8754c3ef09adf98bdd0c3d7d3b8a358a71dc2aebb7105
+size 810702192
diff --git a/checkpoint-300/optimizer.pt b/checkpoint-300/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..466a128a43111e41362a14d5f8536cd5f16ea6da
--- /dev/null
+++ b/checkpoint-300/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16ce8a7a6f6183711a043b0ec65b90fcae17c5eedf8b9d455994abc12e63bb45
+size 1621580026
diff --git a/checkpoint-300/rng_state.pth b/checkpoint-300/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..02e6a75cccce6d8b138f83bf61f04a2d82786ce4
--- /dev/null
+++ b/checkpoint-300/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:374644995be79c967030295cf4e5929f6a33d3b1197b8855bc39718b44686552
+size 14244
diff --git a/checkpoint-300/scheduler.pt b/checkpoint-300/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..743165daadcce85dd53f776ffb1ddebfff2df0c8
--- /dev/null
+++ b/checkpoint-300/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97150327a5f2cf78296d100e337e3afcfcbdfef6d10b94b6a3604a6a856e0e12
+size 1064
diff --git a/checkpoint-300/special_tokens_map.json b/checkpoint-300/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..2a565ac6033dc9066dd29eadd1019f36f9709c86
--- /dev/null
+++ b/checkpoint-300/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57491904f8680d4b52ed440f1f7ba48cad1c31ecf3eb453b03484e6ff4723ae8
+size 473
diff --git a/checkpoint-300/tokenizer.json b/checkpoint-300/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..614d0562ad8d2db17f5aa36ec866a872f7d821a0
--- /dev/null
+++ b/checkpoint-300/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:870f4e2baa6b683221fa52004d5d6f40ab8c9d31961617304b78c910c2c3caf2
+size 2114042
diff --git a/checkpoint-300/tokenizer_config.json b/checkpoint-300/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbc1d807e6afd67bc165abeb593b1c257efad4b3
--- /dev/null
+++ b/checkpoint-300/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4f587ac3a67c3964d3e84248f95ffc8e4f0e64d9de677692df39aa35eb2d11
+size 4845
diff --git a/checkpoint-300/trainer_state.json b/checkpoint-300/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5e32838ee2b77667f83a0711aaaf103993c71278
--- /dev/null
+++ b/checkpoint-300/trainer_state.json
@@ -0,0 +1,2403 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.7774538386783285,
+  "eval_steps": 10,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025915127955944283,
+      "grad_norm": 12.625,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 1.7977,
+      "step": 1
+    },
+    {
+      "epoch": 0.0051830255911888565,
+      "grad_norm": 12.0625,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 1.7787,
+      "step": 2
+    },
+    {
+      "epoch": 0.007774538386783284,
+      "grad_norm": 13.0,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 1.8003,
+      "step": 3
+    },
+    {
+      "epoch": 0.010366051182377713,
+      "grad_norm": 12.5625,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 1.787,
+      "step": 4
+    },
+    {
+      "epoch": 0.01295756397797214,
+      "grad_norm": 12.25,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.7684,
+      "step": 5
+    },
+    {
+      "epoch": 0.015549076773566569,
+      "grad_norm": 12.625,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 1.7544,
+      "step": 6
+    },
+    {
+      "epoch": 0.018140589569160998,
+      "grad_norm": 12.5,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 1.7769,
+      "step": 7
+    },
+    {
+      "epoch": 0.020732102364755426,
+      "grad_norm": 11.4375,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 1.7161,
+      "step": 8
+    },
+    {
+      "epoch": 0.023323615160349854,
+      "grad_norm": 11.5,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 1.7165,
+      "step": 9
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "grad_norm": 12.1875,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.7437,
+      "step": 10
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "eval_loss": 1.7256978750228882,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5814,
+      "eval_samples_per_second": 105.893,
+      "eval_steps_per_second": 13.262,
+      "step": 10
+    },
+    {
+      "epoch": 0.02850664075153871,
+      "grad_norm": 11.4375,
+      "learning_rate": 4.4e-06,
+      "loss": 1.7068,
+      "step": 11
+    },
+    {
+      "epoch": 0.031098153547133137,
+      "grad_norm": 11.75,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 1.688,
+      "step": 12
+    },
+    {
+      "epoch": 0.03368966634272757,
+      "grad_norm": 11.6875,
+      "learning_rate": 5.2e-06,
+      "loss": 1.6602,
+      "step": 13
+    },
+    {
+      "epoch": 0.036281179138321996,
+      "grad_norm": 11.5625,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 1.6389,
+      "step": 14
+    },
+    {
+      "epoch": 0.038872691933916424,
+      "grad_norm": 10.75,
+      "learning_rate": 6e-06,
+      "loss": 1.6228,
+      "step": 15
+    },
+    {
+      "epoch": 0.04146420472951085,
+      "grad_norm": 10.8125,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 1.5823,
+      "step": 16
+    },
+    {
+      "epoch": 0.04405571752510528,
+      "grad_norm": 9.5625,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 1.5134,
+      "step": 17
+    },
+    {
+      "epoch": 0.04664723032069971,
+      "grad_norm": 9.125,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 1.4833,
+      "step": 18
+    },
+    {
+      "epoch": 0.049238743116294136,
+      "grad_norm": 7.375,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 1.4966,
+      "step": 19
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "grad_norm": 7.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.4653,
+      "step": 20
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "eval_loss": 1.4041186571121216,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6226,
+      "eval_samples_per_second": 105.716,
+      "eval_steps_per_second": 13.24,
+      "step": 20
+    },
+    {
+      "epoch": 0.05442176870748299,
+      "grad_norm": 6.0625,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 1.3844,
+      "step": 21
+    },
+    {
+      "epoch": 0.05701328150307742,
+      "grad_norm": 5.375,
+      "learning_rate": 8.8e-06,
+      "loss": 1.3735,
+      "step": 22
+    },
+    {
+      "epoch": 0.05960479429867185,
+      "grad_norm": 5.09375,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 1.3281,
+      "step": 23
+    },
+    {
+      "epoch": 0.062196307094266275,
+      "grad_norm": 5.25,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 1.2837,
+      "step": 24
+    },
+    {
+      "epoch": 0.06478781988986071,
+      "grad_norm": 5.25,
+      "learning_rate": 1e-05,
+      "loss": 1.2777,
+      "step": 25
+    },
+    {
+      "epoch": 0.06737933268545514,
+      "grad_norm": 5.03125,
+      "learning_rate": 1.04e-05,
+      "loss": 1.2465,
+      "step": 26
+    },
+    {
+      "epoch": 0.06997084548104957,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.0800000000000002e-05,
+      "loss": 1.2041,
+      "step": 27
+    },
+    {
+      "epoch": 0.07256235827664399,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.1200000000000001e-05,
+      "loss": 1.1524,
+      "step": 28
+    },
+    {
+      "epoch": 0.07515387107223842,
+      "grad_norm": 4.65625,
+      "learning_rate": 1.16e-05,
+      "loss": 1.1491,
+      "step": 29
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "grad_norm": 5.125,
+      "learning_rate": 1.2e-05,
+      "loss": 1.127,
+      "step": 30
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "eval_loss": 1.105301856994629,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5195,
+      "eval_samples_per_second": 106.161,
+      "eval_steps_per_second": 13.296,
+      "step": 30
+    },
+    {
+      "epoch": 0.08033689666342728,
+      "grad_norm": 5.09375,
+      "learning_rate": 1.2400000000000002e-05,
+      "loss": 1.0993,
+      "step": 31
+    },
+    {
+      "epoch": 0.0829284094590217,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.2800000000000001e-05,
+      "loss": 1.0637,
+      "step": 32
+    },
+    {
+      "epoch": 0.08551992225461613,
+      "grad_norm": 4.75,
+      "learning_rate": 1.3200000000000002e-05,
+      "loss": 1.0447,
+      "step": 33
+    },
+    {
+      "epoch": 0.08811143505021056,
+      "grad_norm": 5.21875,
+      "learning_rate": 1.3600000000000002e-05,
+      "loss": 1.047,
+      "step": 34
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.4e-05,
+      "loss": 0.9841,
+      "step": 35
+    },
+    {
+      "epoch": 0.09329446064139942,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.4400000000000001e-05,
+      "loss": 0.9735,
+      "step": 36
+    },
+    {
+      "epoch": 0.09588597343699384,
+      "grad_norm": 5.71875,
+      "learning_rate": 1.48e-05,
+      "loss": 0.9409,
+      "step": 37
+    },
+    {
+      "epoch": 0.09847748623258827,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.5200000000000002e-05,
+      "loss": 0.9108,
+      "step": 38
+    },
+    {
+      "epoch": 0.1010689990281827,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.5600000000000003e-05,
+      "loss": 0.9029,
+      "step": 39
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.8938,
+      "step": 40
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "eval_loss": 0.8692858815193176,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5016,
+      "eval_samples_per_second": 106.238,
+      "eval_steps_per_second": 13.305,
+      "step": 40
+    },
+    {
+      "epoch": 0.10625202461937155,
+      "grad_norm": 5.125,
+      "learning_rate": 1.64e-05,
+      "loss": 0.8501,
+      "step": 41
+    },
+    {
+      "epoch": 0.10884353741496598,
+      "grad_norm": 4.53125,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 0.8545,
+      "step": 42
+    },
+    {
+      "epoch": 0.11143505021056041,
+      "grad_norm": 5.1875,
+      "learning_rate": 1.72e-05,
+      "loss": 0.8169,
+      "step": 43
+    },
+    {
+      "epoch": 0.11402656300615484,
+      "grad_norm": 4.5,
+      "learning_rate": 1.76e-05,
+      "loss": 0.8102,
+      "step": 44
+    },
+    {
+      "epoch": 0.11661807580174927,
+      "grad_norm": 4.3125,
+      "learning_rate": 1.8e-05,
+      "loss": 0.8093,
+      "step": 45
+    },
+    {
+      "epoch": 0.1192095885973437,
+      "grad_norm": 4.28125,
+      "learning_rate": 1.8400000000000003e-05,
+      "loss": 0.7798,
+      "step": 46
+    },
+    {
+      "epoch": 0.12180110139293812,
+      "grad_norm": 4.09375,
+      "learning_rate": 1.88e-05,
+      "loss": 0.7761,
+      "step": 47
+    },
+    {
+      "epoch": 0.12439261418853255,
+      "grad_norm": 4.125,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.7647,
+      "step": 48
+    },
+    {
+      "epoch": 0.12698412698412698,
+      "grad_norm": 3.765625,
+      "learning_rate": 1.9600000000000002e-05,
+      "loss": 0.7612,
+      "step": 49
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "grad_norm": 4.0,
+      "learning_rate": 2e-05,
+      "loss": 0.7498,
+      "step": 50
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "eval_loss": 0.744697093963623,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6181,
+      "eval_samples_per_second": 105.735,
+      "eval_steps_per_second": 13.242,
+      "step": 50
+    },
+    {
+      "epoch": 0.13216715257531583,
+      "grad_norm": 4.5,
+      "learning_rate": 1.9999560279257315e-05,
+      "loss": 0.7504,
+      "step": 51
+    },
+    {
+      "epoch": 0.13475866537091027,
+      "grad_norm": 3.796875,
+      "learning_rate": 1.999824115570012e-05,
+      "loss": 0.7181,
+      "step": 52
+    },
+    {
+      "epoch": 0.1373501781665047,
+      "grad_norm": 3.703125,
+      "learning_rate": 1.9996042745337617e-05,
+      "loss": 0.7195,
+      "step": 53
+    },
+    {
+      "epoch": 0.13994169096209913,
+      "grad_norm": 3.921875,
+      "learning_rate": 1.999296524150713e-05,
+      "loss": 0.7232,
+      "step": 54
+    },
+    {
+      "epoch": 0.14253320375769354,
+      "grad_norm": 3.6875,
+      "learning_rate": 1.9989008914857115e-05,
+      "loss": 0.725,
+      "step": 55
+    },
+    {
+      "epoch": 0.14512471655328799,
+      "grad_norm": 3.734375,
+      "learning_rate": 1.9984174113323352e-05,
+      "loss": 0.6972,
+      "step": 56
+    },
+    {
+      "epoch": 0.1477162293488824,
+      "grad_norm": 3.5,
+      "learning_rate": 1.9978461262098342e-05,
+      "loss": 0.6792,
+      "step": 57
+    },
+    {
+      "epoch": 0.15030774214447684,
+      "grad_norm": 4.0625,
+      "learning_rate": 1.9971870863593924e-05,
+      "loss": 0.7192,
+      "step": 58
+    },
+    {
+      "epoch": 0.15289925494007126,
+      "grad_norm": 3.75,
+      "learning_rate": 1.9964403497397086e-05,
+      "loss": 0.7107,
+      "step": 59
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9956059820218982e-05,
+      "loss": 0.6866,
+      "step": 60
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "eval_loss": 0.6849498152732849,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5203,
+      "eval_samples_per_second": 106.157,
+      "eval_steps_per_second": 13.295,
+      "step": 60
+    },
+    {
+      "epoch": 0.1580822805312601,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9946840565837206e-05,
+      "loss": 0.6818,
+      "step": 61
+    },
+    {
+      "epoch": 0.16067379332685455,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.9936746545031225e-05,
+      "loss": 0.6855,
+      "step": 62
+    },
+    {
+      "epoch": 0.16326530612244897,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9925778645511112e-05,
+      "loss": 0.6799,
+      "step": 63
+    },
+    {
+      "epoch": 0.1658568189180434,
+      "grad_norm": 3.828125,
+      "learning_rate": 1.991393783183945e-05,
+      "loss": 0.6719,
+      "step": 64
+    },
+    {
+      "epoch": 0.16844833171363785,
+      "grad_norm": 3.0,
+      "learning_rate": 1.990122514534651e-05,
+      "loss": 0.66,
+      "step": 65
+    },
+    {
+      "epoch": 0.17103984450923226,
+      "grad_norm": 2.96875,
+      "learning_rate": 1.988764170403869e-05,
+      "loss": 0.6718,
+      "step": 66
+    },
+    {
+      "epoch": 0.1736313573048267,
+      "grad_norm": 3.15625,
+      "learning_rate": 1.9873188702500163e-05,
+      "loss": 0.6661,
+      "step": 67
+    },
+    {
+      "epoch": 0.17622287010042112,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.9857867411787847e-05,
+      "loss": 0.6523,
+      "step": 68
+    },
+    {
+      "epoch": 0.17881438289601556,
+      "grad_norm": 3.203125,
+      "learning_rate": 1.9841679179319605e-05,
+      "loss": 0.6494,
+      "step": 69
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 2.875,
+      "learning_rate": 1.982462542875576e-05,
+      "loss": 0.6411,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "eval_loss": 0.6511064171791077,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5304,
+      "eval_samples_per_second": 106.113,
+      "eval_steps_per_second": 13.29,
+      "step": 70
+    },
+    {
+      "epoch": 0.18399740848720442,
+      "grad_norm": 3.296875,
+      "learning_rate": 1.9806707659873886e-05,
+      "loss": 0.6463,
+      "step": 71
+    },
+    {
+      "epoch": 0.18658892128279883,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9787927448436913e-05,
+      "loss": 0.6375,
+      "step": 72
+    },
+    {
+      "epoch": 0.18918043407839327,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9768286446054533e-05,
+      "loss": 0.6386,
+      "step": 73
+    },
+    {
+      "epoch": 0.1917719468739877,
+      "grad_norm": 2.875,
+      "learning_rate": 1.974778638003799e-05,
+      "loss": 0.6368,
+      "step": 74
+    },
+    {
+      "epoch": 0.19436345966958213,
+      "grad_norm": 2.953125,
+      "learning_rate": 1.972642905324813e-05,
+      "loss": 0.6271,
+      "step": 75
+    },
+    {
+      "epoch": 0.19695497246517654,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9704216343936875e-05,
+      "loss": 0.6501,
+      "step": 76
+    },
+    {
+      "epoch": 0.19954648526077098,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9681150205582025e-05,
+      "loss": 0.6111,
+      "step": 77
+    },
+    {
+      "epoch": 0.2021379980563654,
+      "grad_norm": 3.171875,
+      "learning_rate": 1.9657232666715485e-05,
+      "loss": 0.651,
+      "step": 78
+    },
+    {
+      "epoch": 0.20472951085195984,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9632465830744846e-05,
+      "loss": 0.6184,
+      "step": 79
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "grad_norm": 3.25,
+      "learning_rate": 1.9606851875768404e-05,
+      "loss": 0.6259,
+      "step": 80
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "eval_loss": 0.6278263330459595,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4892,
+      "eval_samples_per_second": 106.292,
+      "eval_steps_per_second": 13.312,
+      "step": 80
+    },
+    {
+      "epoch": 0.2099125364431487,
+      "grad_norm": 2.875,
+      "learning_rate": 1.9580393054383622e-05,
+      "loss": 0.6288,
+      "step": 81
+    },
+    {
+      "epoch": 0.2125040492387431,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9553091693489018e-05,
+      "loss": 0.6403,
+      "step": 82
+    },
+    {
+      "epoch": 0.21509556203433755,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9524950194079534e-05,
+      "loss": 0.621,
+      "step": 83
+    },
+    {
+      "epoch": 0.21768707482993196,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9495971031035367e-05,
+      "loss": 0.6135,
+      "step": 84
+    },
+    {
+      "epoch": 0.2202785876255264,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9466156752904344e-05,
+      "loss": 0.6211,
+      "step": 85
+    },
+    {
+      "epoch": 0.22287010042112082,
+      "grad_norm": 3.109375,
+      "learning_rate": 1.9435509981677764e-05,
+      "loss": 0.6198,
+      "step": 86
+    },
+    {
+      "epoch": 0.22546161321671526,
+      "grad_norm": 2.625,
+      "learning_rate": 1.9404033412559828e-05,
+      "loss": 0.5972,
+      "step": 87
+    },
+    {
+      "epoch": 0.22805312601230968,
+      "grad_norm": 3.0,
+      "learning_rate": 1.9371729813730606e-05,
+      "loss": 0.614,
+      "step": 88
+    },
+    {
+      "epoch": 0.23064463880790412,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9338602026102594e-05,
+      "loss": 0.5998,
+      "step": 89
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.9304652963070868e-05,
+      "loss": 0.6181,
+      "step": 90
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "eval_loss": 0.6108965277671814,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4982,
+      "eval_samples_per_second": 106.253,
+      "eval_steps_per_second": 13.307,
+      "step": 90
+    },
+    {
+      "epoch": 0.23582766439909297,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.9269885610256866e-05,
+      "loss": 0.6059,
+      "step": 91
+    },
+    {
+      "epoch": 0.2384191771946874,
+      "grad_norm": 2.734375,
+      "learning_rate": 1.9234303025245835e-05,
+      "loss": 0.6165,
+      "step": 92
+    },
+    {
+      "epoch": 0.24101068999028183,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.919790833731791e-05,
+      "loss": 0.6241,
+      "step": 93
+    },
+    {
+      "epoch": 0.24360220278587624,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9160704747172935e-05,
+      "loss": 0.5913,
+      "step": 94
+    },
+    {
+      "epoch": 0.24619371558147068,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9122695526648968e-05,
+      "loss": 0.585,
+      "step": 95
+    },
+    {
+      "epoch": 0.2487852283770651,
+      "grad_norm": 3.484375,
+      "learning_rate": 1.9083884018434545e-05,
+      "loss": 0.5897,
+      "step": 96
+    },
+    {
+      "epoch": 0.2513767411726595,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9044273635774707e-05,
+      "loss": 0.6032,
+      "step": 97
+    },
+    {
+      "epoch": 0.25396825396825395,
+      "grad_norm": 2.75,
+      "learning_rate": 1.900386786217083e-05,
+      "loss": 0.594,
+      "step": 98
+    },
+    {
+      "epoch": 0.2565597667638484,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.8962670251074276e-05,
+      "loss": 0.5958,
+      "step": 99
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.8920684425573865e-05,
+      "loss": 0.5821,
+      "step": 100
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "eval_loss": 0.5988509058952332,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4884,
+      "eval_samples_per_second": 106.295,
+      "eval_steps_per_second": 13.312,
+      "step": 100
+    },
+    {
+      "epoch": 0.2617427923550373,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.887791407807728e-05,
+      "loss": 0.5904,
+      "step": 101
+    },
+    {
+      "epoch": 0.26433430515063167,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.883436296998631e-05,
+      "loss": 0.5913,
+      "step": 102
+    },
+    {
+      "epoch": 0.2669258179462261,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.8790034931366073e-05,
+      "loss": 0.6114,
+      "step": 103
+    },
+    {
+      "epoch": 0.26951733074182055,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.8744933860608183e-05,
+      "loss": 0.5803,
+      "step": 104
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 3.40625,
+      "learning_rate": 1.8699063724087905e-05,
+      "loss": 0.5865,
+      "step": 105
+    },
+    {
+      "epoch": 0.2747003563330094,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.865242855581534e-05,
+      "loss": 0.586,
+      "step": 106
+    },
+    {
+      "epoch": 0.2772918691286038,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.8605032457080655e-05,
+      "loss": 0.5926,
+      "step": 107
+    },
+    {
+      "epoch": 0.27988338192419826,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.8556879596093393e-05,
+      "loss": 0.5927,
+      "step": 108
+    },
+    {
+      "epoch": 0.2824748947197927,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.850797420761592e-05,
+      "loss": 0.5937,
+      "step": 109
+    },
+    {
+      "epoch": 0.2850664075153871,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.8458320592590976e-05,
+      "loss": 0.5644,
+      "step": 110
+    },
+    {
+      "epoch": 0.2850664075153871,
+      "eval_loss": 0.5878391265869141,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.539,
+      "eval_samples_per_second": 106.076,
+      "eval_steps_per_second": 13.285,
+      "step": 110
+    },
+    {
+      "epoch": 0.28765792031098153,
+      "grad_norm": 2.75,
+      "learning_rate": 1.840792311776346e-05,
+      "loss": 0.5783,
+      "step": 111
+    },
+    {
+      "epoch": 0.29024943310657597,
+      "grad_norm": 2.375,
+      "learning_rate": 1.8356786215296387e-05,
+      "loss": 0.5813,
+      "step": 112
+    },
+    {
+      "epoch": 0.2928409459021704,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.83049143823811e-05,
+      "loss": 0.5788,
+      "step": 113
+    },
+    {
+      "epoch": 0.2954324586977648,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.825231218084178e-05,
+      "loss": 0.5831,
+      "step": 114
+    },
+    {
+      "epoch": 0.29802397149335924,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.8198984236734246e-05,
+      "loss": 0.5715,
+      "step": 115
+    },
+    {
+      "epoch": 0.3006154842889537,
+      "grad_norm": 2.546875,
+      "learning_rate": 1.8144935239939147e-05,
+      "loss": 0.5938,
+      "step": 116
+    },
+    {
+      "epoch": 0.3032069970845481,
+      "grad_norm": 2.90625,
+      "learning_rate": 1.8090169943749477e-05,
+      "loss": 0.5911,
+      "step": 117
+    },
+    {
+      "epoch": 0.3057985098801425,
+      "grad_norm": 2.859375,
+      "learning_rate": 1.8034693164452578e-05,
+      "loss": 0.5815,
+      "step": 118
+    },
+    {
+      "epoch": 0.30839002267573695,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.7978509780906578e-05,
+      "loss": 0.5804,
+      "step": 119
+    },
+    {
+      "epoch": 0.3109815354713314,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.7921624734111292e-05,
+      "loss": 0.5749,
+      "step": 120
+    },
+    {
+      "epoch": 0.3109815354713314,
+      "eval_loss": 0.5794549584388733,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5897,
+      "eval_samples_per_second": 105.857,
+      "eval_steps_per_second": 13.258,
+      "step": 120
+    },
+    {
+      "epoch": 0.31357304826692584,
+      "grad_norm": 2.703125,
+      "learning_rate": 1.786404302677374e-05,
+      "loss": 0.5873,
+      "step": 121
+    },
+    {
+      "epoch": 0.3161645610625202,
+      "grad_norm": 2.375,
+      "learning_rate": 1.7805769722868132e-05,
+      "loss": 0.5856,
+      "step": 122
+    },
+    {
+      "epoch": 0.31875607385811466,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.7746809947190567e-05,
+      "loss": 0.578,
+      "step": 123
+    },
+    {
+      "epoch": 0.3213475866537091,
+      "grad_norm": 3.421875,
+      "learning_rate": 1.768716888490832e-05,
+      "loss": 0.5566,
+      "step": 124
+    },
+    {
+      "epoch": 0.32393909944930355,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.762685178110382e-05,
+      "loss": 0.5638,
+      "step": 125
+    },
+    {
+      "epoch": 0.32653061224489793,
+      "grad_norm": 3.125,
+      "learning_rate": 1.7565863940313416e-05,
+      "loss": 0.5827,
+      "step": 126
+    },
+    {
+      "epoch": 0.3291221250404924,
+      "grad_norm": 3.1875,
+      "learning_rate": 1.7504210726060828e-05,
+      "loss": 0.5596,
+      "step": 127
+    },
+    {
+      "epoch": 0.3317136378360868,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.7441897560385493e-05,
+      "loss": 0.5768,
+      "step": 128
+    },
+    {
+      "epoch": 0.33430515063168126,
+      "grad_norm": 2.546875,
+      "learning_rate": 1.7378929923365706e-05,
+      "loss": 0.5732,
+      "step": 129
+    },
+    {
+      "epoch": 0.3368966634272757,
+      "grad_norm": 2.78125,
+      "learning_rate": 1.731531335263669e-05,
+      "loss": 0.5781,
+      "step": 130
+    },
+    {
+      "epoch": 0.3368966634272757,
+      "eval_loss": 0.572753369808197,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5459,
+      "eval_samples_per_second": 106.046,
+      "eval_steps_per_second": 13.281,
+      "step": 130
+    },
+    {
+      "epoch": 0.3394881762228701,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.7251053442903597e-05,
+      "loss": 0.5689,
+      "step": 131
+    },
+    {
+      "epoch": 0.34207968901846453,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.7186155845449465e-05,
+      "loss": 0.5695,
+      "step": 132
+    },
+    {
+      "epoch": 0.34467120181405897,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.712062626763825e-05,
+      "loss": 0.5627,
+      "step": 133
+    },
+    {
+      "epoch": 0.3472627146096534,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.7054470472412876e-05,
+      "loss": 0.5759,
+      "step": 134
+    },
+    {
+      "epoch": 0.3498542274052478,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.698769427778842e-05,
+      "loss": 0.5662,
+      "step": 135
+    },
+    {
+      "epoch": 0.35244574020084224,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.6920303556340462e-05,
+      "loss": 0.5671,
+      "step": 136
+    },
+    {
+      "epoch": 0.3550372529964367,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.6852304234688627e-05,
+      "loss": 0.5572,
+      "step": 137
+    },
+    {
+      "epoch": 0.3576287657920311,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.678370229297535e-05,
+      "loss": 0.5796,
+      "step": 138
+    },
+    {
+      "epoch": 0.3602202785876255,
+      "grad_norm": 3.03125,
+      "learning_rate": 1.6714503764339986e-05,
+      "loss": 0.5642,
+      "step": 139
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.664471473438822e-05,
+      "loss": 0.5711,
+      "step": 140
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "eval_loss": 0.566669762134552,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5724,
+      "eval_samples_per_second": 105.932,
+      "eval_steps_per_second": 13.267,
+      "step": 140
+    },
+    {
+      "epoch": 0.3654033041788144,
+      "grad_norm": 2.625,
+      "learning_rate": 1.657434134065686e-05,
+      "loss": 0.5559,
+      "step": 141
+    },
+    {
+      "epoch": 0.36799481697440883,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.6503389772074104e-05,
+      "loss": 0.575,
+      "step": 142
+    },
+    {
+      "epoch": 0.3705863297700032,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.643186626841524e-05,
+      "loss": 0.5851,
+      "step": 143
+    },
+    {
+      "epoch": 0.37317784256559766,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.6359777119753887e-05,
+      "loss": 0.559,
+      "step": 144
+    },
+    {
+      "epoch": 0.3757693553611921,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.628712866590885e-05,
+      "loss": 0.5759,
+      "step": 145
+    },
+    {
+      "epoch": 0.37836086815678655,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.6213927295886547e-05,
+      "loss": 0.5579,
+      "step": 146
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.6140179447319134e-05,
+      "loss": 0.5717,
+      "step": 147
+    },
+    {
+      "epoch": 0.3835438937479754,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.6065891605898358e-05,
+      "loss": 0.5531,
+      "step": 148
+    },
+    {
+      "epoch": 0.3861354065435698,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.5991070304805184e-05,
+      "loss": 0.5358,
+      "step": 149
+    },
+    {
+      "epoch": 0.38872691933916426,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.5915722124135227e-05,
+      "loss": 0.5715,
+      "step": 150
+    },
+    {
+      "epoch": 0.38872691933916426,
+      "eval_loss": 0.5618191361427307,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5943,
+      "eval_samples_per_second": 105.838,
+      "eval_steps_per_second": 13.255,
+      "step": 150
+    },
+    {
+      "epoch": 0.39131843213475864,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.5839853690320074e-05,
+      "loss": 0.5413,
+      "step": 151
+    },
+    {
+      "epoch": 0.3939099449303531,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.5763471675544547e-05,
+      "loss": 0.5652,
+      "step": 152
+    },
+    {
+      "epoch": 0.3965014577259475,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.5686582797159892e-05,
+      "loss": 0.5682,
+      "step": 153
+    },
+    {
+      "epoch": 0.39909297052154197,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.5609193817093057e-05,
+      "loss": 0.5577,
+      "step": 154
+    },
+    {
+      "epoch": 0.40168448331713635,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.5531311541251995e-05,
+      "loss": 0.5412,
+      "step": 155
+    },
+    {
+      "epoch": 0.4042759961127308,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.5452942818927144e-05,
+      "loss": 0.5441,
+      "step": 156
+    },
+    {
+      "epoch": 0.40686750890832524,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.5374094542189055e-05,
+      "loss": 0.5764,
+      "step": 157
+    },
+    {
+      "epoch": 0.4094590217039197,
+      "grad_norm": 3.078125,
+      "learning_rate": 1.52947736452823e-05,
+      "loss": 0.5763,
+      "step": 158
+    },
+    {
+      "epoch": 0.41205053449951407,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.521498710401561e-05,
+      "loss": 0.5482,
+      "step": 159
+    },
+    {
+      "epoch": 0.4146420472951085,
+      "grad_norm": 2.375,
+      "learning_rate": 1.513474193514842e-05,
+      "loss": 0.559,
+      "step": 160
+    },
+    {
+      "epoch": 0.4146420472951085,
+      "eval_loss": 0.5573883056640625,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5874,
+      "eval_samples_per_second": 105.867,
+      "eval_steps_per_second": 13.259,
+      "step": 160
+    },
+    {
+      "epoch": 0.41723356009070295,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.505404519577379e-05,
+      "loss": 0.5642,
+      "step": 161
+    },
+    {
+      "epoch": 0.4198250728862974,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.4972903982697744e-05,
+      "loss": 0.5535,
+      "step": 162
+    },
+    {
+      "epoch": 0.42241658568189183,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.4891325431815182e-05,
+      "loss": 0.5519,
+      "step": 163
+    },
+    {
+      "epoch": 0.4250080984774862,
+      "grad_norm": 2.0,
+      "learning_rate": 1.48093167174823e-05,
+      "loss": 0.5611,
+      "step": 164
+    },
+    {
+      "epoch": 0.42759961127308066,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.4726885051885654e-05,
+      "loss": 0.5556,
+      "step": 165
+    },
+    {
+      "epoch": 0.4301911240686751,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.4644037684407883e-05,
+      "loss": 0.5289,
+      "step": 166
+    },
+    {
+      "epoch": 0.43278263686426954,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.4560781900990187e-05,
+      "loss": 0.5538,
+      "step": 167
+    },
+    {
+      "epoch": 0.43537414965986393,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.4477125023491536e-05,
+      "loss": 0.5589,
+      "step": 168
+    },
+    {
+      "epoch": 0.43796566245545837,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.4393074409044802e-05,
+      "loss": 0.5423,
+      "step": 169
+    },
+    {
+      "epoch": 0.4405571752510528,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.4308637449409705e-05,
+      "loss": 0.5593,
+      "step": 170
+    },
+    {
+      "epoch": 0.4405571752510528,
+      "eval_loss": 0.553532063961029,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5188,
+      "eval_samples_per_second": 106.163,
+      "eval_steps_per_second": 13.296,
+      "step": 170
+    },
+    {
+      "epoch": 0.44314868804664725,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.4223821570322762e-05,
+      "loss": 0.5789,
+      "step": 171
+    },
+    {
+      "epoch": 0.44574020084224164,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.4138634230844242e-05,
+      "loss": 0.5693,
+      "step": 172
+    },
+    {
+      "epoch": 0.4483317136378361,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.4053082922702183e-05,
+      "loss": 0.5496,
+      "step": 173
+    },
+    {
+      "epoch": 0.4509232264334305,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.3967175169633537e-05,
+      "loss": 0.5362,
+      "step": 174
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.3880918526722497e-05,
+      "loss": 0.5739,
+      "step": 175
+    },
+    {
+      "epoch": 0.45610625202461935,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.3794320579736084e-05,
+      "loss": 0.5493,
+      "step": 176
+    },
+    {
+      "epoch": 0.4586977648202138,
+      "grad_norm": 2.59375,
+      "learning_rate": 1.3707388944457007e-05,
+      "loss": 0.5665,
+      "step": 177
+    },
+    {
+      "epoch": 0.46128927761580824,
+      "grad_norm": 2.125,
+      "learning_rate": 1.3620131266013912e-05,
+      "loss": 0.5557,
+      "step": 178
+    },
+    {
+      "epoch": 0.4638807904114027,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.3532555218209037e-05,
+      "loss": 0.5402,
+      "step": 179
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "grad_norm": 2.375,
+      "learning_rate": 1.344466850284333e-05,
+      "loss": 0.5625,
+      "step": 180
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "eval_loss": 0.5502737164497375,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6233,
+      "eval_samples_per_second": 105.713,
+      "eval_steps_per_second": 13.239,
+      "step": 180
+    },
+    {
+      "epoch": 0.4690638160025915,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.335647884903915e-05,
+      "loss": 0.5492,
+      "step": 181
+    },
+    {
+      "epoch": 0.47165532879818595,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.3267994012560505e-05,
+      "loss": 0.5493,
+      "step": 182
+    },
+    {
+      "epoch": 0.4742468415937804,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.3179221775131005e-05,
+      "loss": 0.5508,
+      "step": 183
+    },
+    {
+      "epoch": 0.4768383543893748,
+      "grad_norm": 3.34375,
+      "learning_rate": 1.3090169943749475e-05,
+      "loss": 0.5427,
+      "step": 184
+    },
+    {
+      "epoch": 0.4794298671849692,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.300084635000341e-05,
+      "loss": 0.568,
+      "step": 185
+    },
+    {
+      "epoch": 0.48202137998056366,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.2911258849380199e-05,
+      "loss": 0.5478,
+      "step": 186
+    },
+    {
+      "epoch": 0.4846128927761581,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.2821415320576309e-05,
+      "loss": 0.5582,
+      "step": 187
+    },
+    {
+      "epoch": 0.4872044055717525,
+      "grad_norm": 3.0,
+      "learning_rate": 1.273132366480438e-05,
+      "loss": 0.5368,
+      "step": 188
+    },
+    {
+      "epoch": 0.4897959183673469,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.2640991805098369e-05,
+      "loss": 0.5608,
+      "step": 189
+    },
+    {
+      "epoch": 0.49238743116294137,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.2550427685616767e-05,
+      "loss": 0.5482,
+      "step": 190
+    },
+    {
+      "epoch": 0.49238743116294137,
+      "eval_loss": 0.547302782535553,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5429,
+      "eval_samples_per_second": 106.059,
+      "eval_steps_per_second": 13.283,
+      "step": 190
+    },
+    {
+      "epoch": 0.4949789439585358,
+      "grad_norm": 2.640625,
+      "learning_rate": 1.2459639270943944e-05,
+      "loss": 0.5508,
+      "step": 191
+    },
+    {
+      "epoch": 0.4975704567541302,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.2368634545389732e-05,
+      "loss": 0.5402,
+      "step": 192
+    },
+    {
+      "epoch": 0.5001619695497247,
+      "grad_norm": 2.25,
+      "learning_rate": 1.2277421512287227e-05,
+      "loss": 0.5313,
+      "step": 193
+    },
+    {
+      "epoch": 0.502753482345319,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.2186008193288963e-05,
+      "loss": 0.5535,
+      "step": 194
+    },
+    {
+      "epoch": 0.5053449951409135,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.2094402627661447e-05,
+      "loss": 0.5375,
+      "step": 195
+    },
+    {
+      "epoch": 0.5079365079365079,
+      "grad_norm": 2.375,
+      "learning_rate": 1.2002612871578143e-05,
+      "loss": 0.5564,
+      "step": 196
+    },
+    {
+      "epoch": 0.5105280207321024,
+      "grad_norm": 2.5,
+      "learning_rate": 1.1910646997411002e-05,
+      "loss": 0.5459,
+      "step": 197
+    },
+    {
+      "epoch": 0.5131195335276968,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.1818513093020514e-05,
+      "loss": 0.5545,
+      "step": 198
+    },
+    {
+      "epoch": 0.5157110463232912,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.172621926104446e-05,
+      "loss": 0.535,
+      "step": 199
+    },
+    {
+      "epoch": 0.5183025591188857,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.1633773618185302e-05,
+      "loss": 0.5469,
+      "step": 200
+    },
+    {
+      "epoch": 0.5183025591188857,
+      "eval_loss": 0.545219898223877,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4828,
+      "eval_samples_per_second": 106.319,
+      "eval_steps_per_second": 13.315,
+      "step": 200
+    },
+    {
+      "epoch": 0.5208940719144801,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.1541184294496393e-05,
+      "loss": 0.5349,
+      "step": 201
+    },
+    {
+      "epoch": 0.5234855847100746,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.144845943266696e-05,
+      "loss": 0.5271,
+      "step": 202
+    },
+    {
+      "epoch": 0.5260770975056689,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.1355607187306036e-05,
+      "loss": 0.5431,
+      "step": 203
+    },
+    {
+      "epoch": 0.5286686103012633,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.1262635724225273e-05,
+      "loss": 0.5344,
+      "step": 204
+    },
+    {
+      "epoch": 0.5312601230968578,
+      "grad_norm": 2.53125,
+      "learning_rate": 1.1169553219720828e-05,
+      "loss": 0.5386,
+      "step": 205
+    },
+    {
+      "epoch": 0.5338516358924522,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.1076367859854305e-05,
+      "loss": 0.5277,
+      "step": 206
+    },
+    {
+      "epoch": 0.5364431486880467,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.0983087839732834e-05,
+      "loss": 0.5339,
+      "step": 207
+    },
+    {
+      "epoch": 0.5390346614836411,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.0889721362788361e-05,
+      "loss": 0.5535,
+      "step": 208
+    },
+    {
+      "epoch": 0.5416261742792355,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.0796276640056201e-05,
+      "loss": 0.5385,
+      "step": 209
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.070276188945293e-05,
+      "loss": 0.5628,
+      "step": 210
+    },
+    {
+      "epoch": 0.54421768707483,
+      "eval_loss": 0.5431892275810242,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5521,
+      "eval_samples_per_second": 106.019,
+      "eval_steps_per_second": 13.278,
+      "step": 210
+    },
+    {
+      "epoch": 0.5468091998704243,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.060918533505367e-05,
+      "loss": 0.538,
+      "step": 211
+    },
+    {
+      "epoch": 0.5494007126660188,
+      "grad_norm": 3.1875,
+      "learning_rate": 1.0515555206368815e-05,
+      "loss": 0.53,
+      "step": 212
+    },
+    {
+      "epoch": 0.5519922254616132,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.0421879737620312e-05,
+      "loss": 0.5315,
+      "step": 213
+    },
+    {
+      "epoch": 0.5545837382572076,
+      "grad_norm": 2.125,
+      "learning_rate": 1.03281671670175e-05,
+      "loss": 0.5408,
+      "step": 214
+    },
+    {
+      "epoch": 0.5571752510528021,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.0234425736032607e-05,
+      "loss": 0.5436,
+      "step": 215
+    },
+    {
+      "epoch": 0.5597667638483965,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.014066368867596e-05,
+      "loss": 0.5269,
+      "step": 216
+    },
+    {
+      "epoch": 0.562358276643991,
+      "grad_norm": 2.125,
+      "learning_rate": 1.0046889270770987e-05,
+      "loss": 0.5412,
+      "step": 217
+    },
+    {
+      "epoch": 0.5649497894395854,
+      "grad_norm": 2.3125,
+      "learning_rate": 9.953110729229018e-06,
+      "loss": 0.5346,
+      "step": 218
+    },
+    {
+      "epoch": 0.5675413022351797,
+      "grad_norm": 2.109375,
+      "learning_rate": 9.859336311324042e-06,
+      "loss": 0.5435,
+      "step": 219
+    },
+    {
+      "epoch": 0.5701328150307742,
+      "grad_norm": 2.125,
+      "learning_rate": 9.765574263967397e-06,
+      "loss": 0.5482,
+      "step": 220
+    },
+    {
+      "epoch": 0.5701328150307742,
+      "eval_loss": 0.5415154695510864,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.571,
+      "eval_samples_per_second": 105.938,
+      "eval_steps_per_second": 13.268,
+      "step": 220
+    },
+    {
+      "epoch": 0.5727243278263686,
+      "grad_norm": 2.21875,
+      "learning_rate": 9.671832832982502e-06,
+      "loss": 0.5401,
+      "step": 221
+    },
+    {
+      "epoch": 0.5753158406219631,
+      "grad_norm": 2.3125,
+      "learning_rate": 9.57812026237969e-06,
+      "loss": 0.5359,
+      "step": 222
+    },
+    {
+      "epoch": 0.5779073534175575,
+      "grad_norm": 2.375,
+      "learning_rate": 9.484444793631186e-06,
+      "loss": 0.5366,
+      "step": 223
+    },
+    {
+      "epoch": 0.5804988662131519,
+      "grad_norm": 2.09375,
+      "learning_rate": 9.390814664946332e-06,
+      "loss": 0.5343,
+      "step": 224
+    },
+    {
+      "epoch": 0.5830903790087464,
+      "grad_norm": 2.03125,
+      "learning_rate": 9.297238110547075e-06,
+      "loss": 0.5385,
+      "step": 225
+    },
+    {
+      "epoch": 0.5856818918043408,
+      "grad_norm": 2.21875,
+      "learning_rate": 9.203723359943802e-06,
+      "loss": 0.5473,
+      "step": 226
+    },
+    {
+      "epoch": 0.5882734045999352,
+      "grad_norm": 2.53125,
+      "learning_rate": 9.110278637211642e-06,
+      "loss": 0.5316,
+      "step": 227
+    },
+    {
+      "epoch": 0.5908649173955296,
+      "grad_norm": 2.046875,
+      "learning_rate": 9.016912160267168e-06,
+      "loss": 0.53,
+      "step": 228
+    },
+    {
+      "epoch": 0.593456430191124,
+      "grad_norm": 2.4375,
+      "learning_rate": 8.923632140145702e-06,
+      "loss": 0.5497,
+      "step": 229
+    },
+    {
+      "epoch": 0.5960479429867185,
+      "grad_norm": 2.15625,
+      "learning_rate": 8.830446780279175e-06,
+      "loss": 0.5499,
+      "step": 230
+    },
+    {
+      "epoch": 0.5960479429867185,
+      "eval_loss": 0.5400256514549255,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5551,
+      "eval_samples_per_second": 106.007,
+      "eval_steps_per_second": 13.276,
+      "step": 230
+    },
+    {
+      "epoch": 0.5986394557823129,
+      "grad_norm": 2.15625,
+      "learning_rate": 8.73736427577473e-06,
+      "loss": 0.5301,
+      "step": 231
+    },
+    {
+      "epoch": 0.6012309685779074,
+      "grad_norm": 2.171875,
+      "learning_rate": 8.644392812693969e-06,
+      "loss": 0.5542,
+      "step": 232
+    },
+    {
+      "epoch": 0.6038224813735018,
+      "grad_norm": 2.1875,
+      "learning_rate": 8.551540567333041e-06,
+      "loss": 0.5374,
+      "step": 233
+    },
+    {
+      "epoch": 0.6064139941690962,
+      "grad_norm": 2.421875,
+      "learning_rate": 8.458815705503612e-06,
+      "loss": 0.549,
+      "step": 234
+    },
+    {
+      "epoch": 0.6090055069646907,
+      "grad_norm": 2.0625,
+      "learning_rate": 8.366226381814698e-06,
+      "loss": 0.5383,
+      "step": 235
+    },
+    {
+      "epoch": 0.611597019760285,
+      "grad_norm": 2.203125,
+      "learning_rate": 8.273780738955545e-06,
+      "loss": 0.5297,
+      "step": 236
+    },
+    {
+      "epoch": 0.6141885325558795,
+      "grad_norm": 2.140625,
+      "learning_rate": 8.181486906979488e-06,
+      "loss": 0.5303,
+      "step": 237
+    },
+    {
+      "epoch": 0.6167800453514739,
+      "grad_norm": 2.109375,
+      "learning_rate": 8.089353002589003e-06,
+      "loss": 0.5429,
+      "step": 238
+    },
+    {
+      "epoch": 0.6193715581470683,
+      "grad_norm": 2.171875,
+      "learning_rate": 7.997387128421859e-06,
+      "loss": 0.5442,
+      "step": 239
+    },
+    {
+      "epoch": 0.6219630709426628,
+      "grad_norm": 2.328125,
+      "learning_rate": 7.905597372338558e-06,
+      "loss": 0.5451,
+      "step": 240
+    },
+    {
+      "epoch": 0.6219630709426628,
+      "eval_loss": 0.5389755964279175,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5608,
+      "eval_samples_per_second": 105.982,
+      "eval_steps_per_second": 13.273,
+      "step": 240
+    },
+    {
+      "epoch": 0.6245545837382572,
+      "grad_norm": 2.078125,
+      "learning_rate": 7.81399180671104e-06,
+      "loss": 0.5392,
+      "step": 241
+    },
+    {
+      "epoch": 0.6271460965338517,
+      "grad_norm": 2.140625,
+      "learning_rate": 7.722578487712776e-06,
+      "loss": 0.5303,
+      "step": 242
+    },
+    {
+      "epoch": 0.6297376093294461,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.631365454610273e-06,
+      "loss": 0.5387,
+      "step": 243
+    },
+    {
+      "epoch": 0.6323291221250404,
+      "grad_norm": 2.265625,
+      "learning_rate": 7.540360729056059e-06,
+      "loss": 0.5528,
+      "step": 244
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.449572314383237e-06,
+      "loss": 0.5488,
+      "step": 245
+    },
+    {
+      "epoch": 0.6375121477162293,
+      "grad_norm": 2.03125,
+      "learning_rate": 7.359008194901633e-06,
+      "loss": 0.5216,
+      "step": 246
+    },
+    {
+      "epoch": 0.6401036605118238,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.268676335195624e-06,
+      "loss": 0.5314,
+      "step": 247
+    },
+    {
+      "epoch": 0.6426951733074182,
+      "grad_norm": 2.125,
+      "learning_rate": 7.1785846794236945e-06,
+      "loss": 0.5411,
+      "step": 248
+    },
+    {
+      "epoch": 0.6452866861030127,
+      "grad_norm": 2.140625,
+      "learning_rate": 7.088741150619804e-06,
+      "loss": 0.5272,
+      "step": 249
+    },
+    {
+      "epoch": 0.6478781988986071,
+      "grad_norm": 2.125,
+      "learning_rate": 6.999153649996595e-06,
+      "loss": 0.5344,
+      "step": 250
+    },
+    {
+      "epoch": 0.6478781988986071,
+      "eval_loss": 0.538066029548645,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5108,
+      "eval_samples_per_second": 106.198,
+      "eval_steps_per_second": 13.3,
+      "step": 250
+    },
+    {
+      "epoch": 0.6504697116942015,
+      "grad_norm": 2.09375,
+      "learning_rate": 6.909830056250527e-06,
+      "loss": 0.532,
+      "step": 251
+    },
+    {
+      "epoch": 0.6530612244897959,
+      "grad_norm": 2.09375,
+      "learning_rate": 6.820778224868998e-06,
+      "loss": 0.5376,
+      "step": 252
+    },
+    {
+      "epoch": 0.6556527372853903,
+      "grad_norm": 2.15625,
+      "learning_rate": 6.732005987439494e-06,
+      "loss": 0.5501,
+      "step": 253
+    },
+    {
+      "epoch": 0.6582442500809847,
+      "grad_norm": 2.0625,
+      "learning_rate": 6.6435211509608546e-06,
+      "loss": 0.532,
+      "step": 254
+    },
+    {
+      "epoch": 0.6608357628765792,
+      "grad_norm": 2.3125,
+      "learning_rate": 6.555331497156671e-06,
+      "loss": 0.5406,
+      "step": 255
+    },
+    {
+      "epoch": 0.6634272756721736,
+      "grad_norm": 2.03125,
+      "learning_rate": 6.467444781790967e-06,
+      "loss": 0.5249,
+      "step": 256
+    },
+    {
+      "epoch": 0.6660187884677681,
+      "grad_norm": 2.171875,
+      "learning_rate": 6.379868733986089e-06,
+      "loss": 0.5195,
+      "step": 257
+    },
+    {
+      "epoch": 0.6686103012633625,
+      "grad_norm": 2.1875,
+      "learning_rate": 6.292611055542998e-06,
+      "loss": 0.5397,
+      "step": 258
+    },
+    {
+      "epoch": 0.671201814058957,
+      "grad_norm": 2.21875,
+      "learning_rate": 6.205679420263917e-06,
+      "loss": 0.5455,
+      "step": 259
+    },
+    {
+      "epoch": 0.6737933268545514,
+      "grad_norm": 2.09375,
+      "learning_rate": 6.119081473277502e-06,
+      "loss": 0.5414,
+      "step": 260
+    },
+    {
+      "epoch": 0.6737933268545514,
+      "eval_loss": 0.5372032523155212,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5567,
+      "eval_samples_per_second": 106.0,
+      "eval_steps_per_second": 13.275,
+      "step": 260
+    },
+    {
+      "epoch": 0.6763848396501457,
+      "grad_norm": 2.5,
+      "learning_rate": 6.032824830366466e-06,
+      "loss": 0.5398,
+      "step": 261
+    },
+    {
+      "epoch": 0.6789763524457402,
+      "grad_norm": 2.109375,
+      "learning_rate": 5.946917077297819e-06,
+      "loss": 0.5512,
+      "step": 262
+    },
+    {
+      "epoch": 0.6815678652413346,
+      "grad_norm": 2.125,
+      "learning_rate": 5.86136576915576e-06,
+      "loss": 0.5486,
+      "step": 263
+    },
+    {
+      "epoch": 0.6841593780369291,
+      "grad_norm": 2.203125,
+      "learning_rate": 5.7761784296772395e-06,
+      "loss": 0.543,
+      "step": 264
+    },
+    {
+      "epoch": 0.6867508908325235,
+      "grad_norm": 2.171875,
+      "learning_rate": 5.6913625505902966e-06,
+      "loss": 0.5065,
+      "step": 265
+    },
+    {
+      "epoch": 0.6893424036281179,
+      "grad_norm": 2.171875,
+      "learning_rate": 5.6069255909551986e-06,
+      "loss": 0.5328,
+      "step": 266
+    },
+    {
+      "epoch": 0.6919339164237124,
+      "grad_norm": 2.015625,
+      "learning_rate": 5.522874976508464e-06,
+      "loss": 0.5404,
+      "step": 267
+    },
+    {
+      "epoch": 0.6945254292193068,
+      "grad_norm": 1.9375,
+      "learning_rate": 5.439218099009822e-06,
+      "loss": 0.528,
+      "step": 268
+    },
+    {
+      "epoch": 0.6971169420149012,
+      "grad_norm": 2.0625,
+      "learning_rate": 5.355962315592118e-06,
+      "loss": 0.5377,
+      "step": 269
+    },
+    {
+      "epoch": 0.6997084548104956,
+      "grad_norm": 2.078125,
+      "learning_rate": 5.273114948114346e-06,
+      "loss": 0.5361,
+      "step": 270
+    },
+    {
+      "epoch": 0.6997084548104956,
+      "eval_loss": 0.5366343259811401,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5782,
+      "eval_samples_per_second": 105.907,
+      "eval_steps_per_second": 13.264,
+      "step": 270
+    },
+    {
+      "epoch": 0.70229996760609,
+      "grad_norm": 2.203125,
+      "learning_rate": 5.190683282517701e-06,
+      "loss": 0.5295,
+      "step": 271
+    },
+    {
+      "epoch": 0.7048914804016845,
+      "grad_norm": 2.15625,
+      "learning_rate": 5.108674568184822e-06,
+      "loss": 0.5333,
+      "step": 272
+    },
+    {
+      "epoch": 0.7074829931972789,
+      "grad_norm": 2.0,
+      "learning_rate": 5.027096017302261e-06,
+      "loss": 0.525,
+      "step": 273
+    },
+    {
+      "epoch": 0.7100745059928734,
+      "grad_norm": 2.140625,
+      "learning_rate": 4.945954804226214e-06,
+      "loss": 0.5328,
+      "step": 274
+    },
+    {
+      "epoch": 0.7126660187884678,
+      "grad_norm": 1.953125,
+      "learning_rate": 4.865258064851579e-06,
+      "loss": 0.5453,
+      "step": 275
+    },
+    {
+      "epoch": 0.7152575315840622,
+      "grad_norm": 2.109375,
+      "learning_rate": 4.785012895984397e-06,
+      "loss": 0.537,
+      "step": 276
+    },
+    {
+      "epoch": 0.7178490443796566,
+      "grad_norm": 1.9609375,
+      "learning_rate": 4.705226354717704e-06,
+      "loss": 0.5459,
+      "step": 277
+    },
+    {
+      "epoch": 0.720440557175251,
+      "grad_norm": 1.9296875,
+      "learning_rate": 4.625905457810943e-06,
+      "loss": 0.5444,
+      "step": 278
+    },
+    {
+      "epoch": 0.7230320699708455,
+      "grad_norm": 2.25,
+      "learning_rate": 4.547057181072861e-06,
+      "loss": 0.532,
+      "step": 279
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 2.046875,
+      "learning_rate": 4.468688458748006e-06,
+      "loss": 0.5209,
+      "step": 280
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "eval_loss": 0.5361185073852539,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5554,
+      "eval_samples_per_second": 106.005,
+      "eval_steps_per_second": 13.276,
+      "step": 280
+    },
+    {
+      "epoch": 0.7282150955620343,
+      "grad_norm": 1.9375,
+      "learning_rate": 4.390806182906946e-06,
+      "loss": 0.5304,
+      "step": 281
+    },
+    {
+      "epoch": 0.7308066083576288,
+      "grad_norm": 1.9921875,
+      "learning_rate": 4.313417202840106e-06,
+      "loss": 0.5532,
+      "step": 282
+    },
+    {
+      "epoch": 0.7333981211532232,
+      "grad_norm": 2.265625,
+      "learning_rate": 4.2365283244554545e-06,
+      "loss": 0.5358,
+      "step": 283
+    },
+    {
+      "epoch": 0.7359896339488177,
+      "grad_norm": 2.34375,
+      "learning_rate": 4.160146309679928e-06,
+      "loss": 0.5444,
+      "step": 284
+    },
+    {
+      "epoch": 0.738581146744412,
+      "grad_norm": 2.171875,
+      "learning_rate": 4.084277875864776e-06,
+      "loss": 0.5374,
+      "step": 285
+    },
+    {
+      "epoch": 0.7411726595400064,
+      "grad_norm": 2.53125,
+      "learning_rate": 4.008929695194819e-06,
+      "loss": 0.5441,
+      "step": 286
+    },
+    {
+      "epoch": 0.7437641723356009,
+      "grad_norm": 2.71875,
+      "learning_rate": 3.934108394101645e-06,
+      "loss": 0.5383,
+      "step": 287
+    },
+    {
+      "epoch": 0.7463556851311953,
+      "grad_norm": 2.3125,
+      "learning_rate": 3.859820552680867e-06,
+      "loss": 0.5481,
+      "step": 288
+    },
+    {
+      "epoch": 0.7489471979267898,
+      "grad_norm": 2.359375,
+      "learning_rate": 3.786072704113456e-06,
+      "loss": 0.5435,
+      "step": 289
+    },
+    {
+      "epoch": 0.7515387107223842,
+      "grad_norm": 2.0,
+      "learning_rate": 3.712871334091154e-06,
+      "loss": 0.5247,
+      "step": 290
+    },
+    {
+      "epoch": 0.7515387107223842,
+      "eval_loss": 0.5358829498291016,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5374,
+      "eval_samples_per_second": 106.083,
+      "eval_steps_per_second": 13.286,
+      "step": 290
+    },
+    {
+      "epoch": 0.7541302235179786,
+      "grad_norm": 2.03125,
+      "learning_rate": 3.640222880246117e-06,
+      "loss": 0.5215,
+      "step": 291
+    },
+    {
+      "epoch": 0.7567217363135731,
+      "grad_norm": 2.09375,
+      "learning_rate": 3.5681337315847677e-06,
+      "loss": 0.5337,
+      "step": 292
+    },
+    {
+      "epoch": 0.7593132491091675,
+      "grad_norm": 1.9609375,
+      "learning_rate": 3.4966102279258964e-06,
+      "loss": 0.5353,
+      "step": 293
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 1.984375,
+      "learning_rate": 3.425658659343141e-06,
+      "loss": 0.5454,
+      "step": 294
+    },
+    {
+      "epoch": 0.7644962747003563,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.355285265611784e-06,
+      "loss": 0.5172,
+      "step": 295
+    },
+    {
+      "epoch": 0.7670877874959507,
+      "grad_norm": 1.9765625,
+      "learning_rate": 3.285496235660013e-06,
+      "loss": 0.5267,
+      "step": 296
+    },
+    {
+      "epoch": 0.7696793002915452,
+      "grad_norm": 1.9765625,
+      "learning_rate": 3.216297707024655e-06,
+      "loss": 0.5436,
+      "step": 297
+    },
+    {
+      "epoch": 0.7722708130871396,
+      "grad_norm": 2.1875,
+      "learning_rate": 3.147695765311377e-06,
+      "loss": 0.5349,
+      "step": 298
+    },
+    {
+      "epoch": 0.7748623258827341,
+      "grad_norm": 2.0,
+      "learning_rate": 3.079696443659538e-06,
+      "loss": 0.5437,
+      "step": 299
+    },
+    {
+      "epoch": 0.7774538386783285,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.0123057222115835e-06,
+      "loss": 0.5185,
+      "step": 300
+    },
+    {
+      "epoch": 0.7774538386783285,
+      "eval_loss": 0.5356724858283997,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5736,
+      "eval_samples_per_second": 105.927,
+      "eval_steps_per_second": 13.266,
+      "step": 300
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 385,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.3477252800512e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-300/training_args.bin b/checkpoint-300/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f7aa13c9cfd007e1e40a3fab731420d1b2409529
--- /dev/null
+++ b/checkpoint-300/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca15ebb16a59dc055f8ae01b1eb1038e869a6556aa3b7b802c58f2dde6a2993c
+size 5304
diff --git a/checkpoint-385/config.json b/checkpoint-385/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..4a3fd3ad7e0af9bc29478da576fa28f4642ce531
--- /dev/null
+++ b/checkpoint-385/config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322b2d3fc9de58487c02f1cd651d484d73244f9704de3d5feea36d002a38af0f
+size 748
diff --git a/checkpoint-385/generation_config.json b/checkpoint-385/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..475e140691313b73815f90f47ad5fde439467dfd
--- /dev/null
+++ b/checkpoint-385/generation_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82bdd024a0ba026e4001e0433d3922a27b1242913db7be87e2851d17165b99c8
+size 111
diff --git a/checkpoint-385/model.safetensors b/checkpoint-385/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..044e065a9d2f95c14465bea33ae152611e30f650
--- /dev/null
+++ b/checkpoint-385/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:783f210de0e18c0cadd41c5bb8b4d1bbf18230cf2d1546ebf37450c870ceee8d
+size 810702192
diff --git a/checkpoint-385/optimizer.pt b/checkpoint-385/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e844061a74946eee9abfa18ded37944795cb4c69
--- /dev/null
+++ b/checkpoint-385/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5d12293b5ce20e50619268243be579e5275bbe8d306cf25e741072c12661697
+size 1621580026
diff --git a/checkpoint-385/rng_state.pth b/checkpoint-385/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6be9c3c0c806d8f8fdeb98cf4c692bd7ef77965f
--- /dev/null
+++ b/checkpoint-385/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:358ab1e5c417d33e0932059d98ea2c06804bddb23da73a96b2bf00951951b130
+size 14244
diff --git a/checkpoint-385/scheduler.pt b/checkpoint-385/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a16d8a252a7a09251e5fc943efb4bfd59da392ba
--- /dev/null
+++ b/checkpoint-385/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08507fbaf34837c4bf6230303cd0aebaa835a7789433f0db443e7ed81a9694d5
+size 1064
diff --git a/checkpoint-385/special_tokens_map.json b/checkpoint-385/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..2a565ac6033dc9066dd29eadd1019f36f9709c86
--- /dev/null
+++ b/checkpoint-385/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57491904f8680d4b52ed440f1f7ba48cad1c31ecf3eb453b03484e6ff4723ae8
+size 473
diff --git a/checkpoint-385/tokenizer.json b/checkpoint-385/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..614d0562ad8d2db17f5aa36ec866a872f7d821a0
--- /dev/null
+++ b/checkpoint-385/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:870f4e2baa6b683221fa52004d5d6f40ab8c9d31961617304b78c910c2c3caf2
+size 2114042
diff --git a/checkpoint-385/tokenizer_config.json b/checkpoint-385/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbc1d807e6afd67bc165abeb593b1c257efad4b3
--- /dev/null
+++ b/checkpoint-385/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4f587ac3a67c3964d3e84248f95ffc8e4f0e64d9de677692df39aa35eb2d11
+size 4845
diff --git a/checkpoint-385/trainer_state.json b/checkpoint-385/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..633e5525982659cbf570f868c90a9589b705b286
--- /dev/null
+++ b/checkpoint-385/trainer_state.json
@@ -0,0 +1,3070 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9977324263038548,
+  "eval_steps": 10,
+  "global_step": 385,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025915127955944283,
+      "grad_norm": 12.625,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 1.7977,
+      "step": 1
+    },
+    {
+      "epoch": 0.0051830255911888565,
+      "grad_norm": 12.0625,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 1.7787,
+      "step": 2
+    },
+    {
+      "epoch": 0.007774538386783284,
+      "grad_norm": 13.0,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 1.8003,
+      "step": 3
+    },
+    {
+      "epoch": 0.010366051182377713,
+      "grad_norm": 12.5625,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 1.787,
+      "step": 4
+    },
+    {
+      "epoch": 0.01295756397797214,
+      "grad_norm": 12.25,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.7684,
+      "step": 5
+    },
+    {
+      "epoch": 0.015549076773566569,
+      "grad_norm": 12.625,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 1.7544,
+      "step": 6
+    },
+    {
+      "epoch": 0.018140589569160998,
+      "grad_norm": 12.5,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 1.7769,
+      "step": 7
+    },
+    {
+      "epoch": 0.020732102364755426,
+      "grad_norm": 11.4375,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 1.7161,
+      "step": 8
+    },
+    {
+      "epoch": 0.023323615160349854,
+      "grad_norm": 11.5,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 1.7165,
+      "step": 9
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "grad_norm": 12.1875,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.7437,
+      "step": 10
+    },
+    {
+      "epoch": 0.02591512795594428,
+      "eval_loss": 1.7256978750228882,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5814,
+      "eval_samples_per_second": 105.893,
+      "eval_steps_per_second": 13.262,
+      "step": 10
+    },
+    {
+      "epoch": 0.02850664075153871,
+      "grad_norm": 11.4375,
+      "learning_rate": 4.4e-06,
+      "loss": 1.7068,
+      "step": 11
+    },
+    {
+      "epoch": 0.031098153547133137,
+      "grad_norm": 11.75,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 1.688,
+      "step": 12
+    },
+    {
+      "epoch": 0.03368966634272757,
+      "grad_norm": 11.6875,
+      "learning_rate": 5.2e-06,
+      "loss": 1.6602,
+      "step": 13
+    },
+    {
+      "epoch": 0.036281179138321996,
+      "grad_norm": 11.5625,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 1.6389,
+      "step": 14
+    },
+    {
+      "epoch": 0.038872691933916424,
+      "grad_norm": 10.75,
+      "learning_rate": 6e-06,
+      "loss": 1.6228,
+      "step": 15
+    },
+    {
+      "epoch": 0.04146420472951085,
+      "grad_norm": 10.8125,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 1.5823,
+      "step": 16
+    },
+    {
+      "epoch": 0.04405571752510528,
+      "grad_norm": 9.5625,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 1.5134,
+      "step": 17
+    },
+    {
+      "epoch": 0.04664723032069971,
+      "grad_norm": 9.125,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 1.4833,
+      "step": 18
+    },
+    {
+      "epoch": 0.049238743116294136,
+      "grad_norm": 7.375,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 1.4966,
+      "step": 19
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "grad_norm": 7.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.4653,
+      "step": 20
+    },
+    {
+      "epoch": 0.05183025591188856,
+      "eval_loss": 1.4041186571121216,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6226,
+      "eval_samples_per_second": 105.716,
+      "eval_steps_per_second": 13.24,
+      "step": 20
+    },
+    {
+      "epoch": 0.05442176870748299,
+      "grad_norm": 6.0625,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 1.3844,
+      "step": 21
+    },
+    {
+      "epoch": 0.05701328150307742,
+      "grad_norm": 5.375,
+      "learning_rate": 8.8e-06,
+      "loss": 1.3735,
+      "step": 22
+    },
+    {
+      "epoch": 0.05960479429867185,
+      "grad_norm": 5.09375,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 1.3281,
+      "step": 23
+    },
+    {
+      "epoch": 0.062196307094266275,
+      "grad_norm": 5.25,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 1.2837,
+      "step": 24
+    },
+    {
+      "epoch": 0.06478781988986071,
+      "grad_norm": 5.25,
+      "learning_rate": 1e-05,
+      "loss": 1.2777,
+      "step": 25
+    },
+    {
+      "epoch": 0.06737933268545514,
+      "grad_norm": 5.03125,
+      "learning_rate": 1.04e-05,
+      "loss": 1.2465,
+      "step": 26
+    },
+    {
+      "epoch": 0.06997084548104957,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.0800000000000002e-05,
+      "loss": 1.2041,
+      "step": 27
+    },
+    {
+      "epoch": 0.07256235827664399,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.1200000000000001e-05,
+      "loss": 1.1524,
+      "step": 28
+    },
+    {
+      "epoch": 0.07515387107223842,
+      "grad_norm": 4.65625,
+      "learning_rate": 1.16e-05,
+      "loss": 1.1491,
+      "step": 29
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "grad_norm": 5.125,
+      "learning_rate": 1.2e-05,
+      "loss": 1.127,
+      "step": 30
+    },
+    {
+      "epoch": 0.07774538386783285,
+      "eval_loss": 1.105301856994629,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5195,
+      "eval_samples_per_second": 106.161,
+      "eval_steps_per_second": 13.296,
+      "step": 30
+    },
+    {
+      "epoch": 0.08033689666342728,
+      "grad_norm": 5.09375,
+      "learning_rate": 1.2400000000000002e-05,
+      "loss": 1.0993,
+      "step": 31
+    },
+    {
+      "epoch": 0.0829284094590217,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.2800000000000001e-05,
+      "loss": 1.0637,
+      "step": 32
+    },
+    {
+      "epoch": 0.08551992225461613,
+      "grad_norm": 4.75,
+      "learning_rate": 1.3200000000000002e-05,
+      "loss": 1.0447,
+      "step": 33
+    },
+    {
+      "epoch": 0.08811143505021056,
+      "grad_norm": 5.21875,
+      "learning_rate": 1.3600000000000002e-05,
+      "loss": 1.047,
+      "step": 34
+    },
+    {
+      "epoch": 0.09070294784580499,
+      "grad_norm": 4.9375,
+      "learning_rate": 1.4e-05,
+      "loss": 0.9841,
+      "step": 35
+    },
+    {
+      "epoch": 0.09329446064139942,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.4400000000000001e-05,
+      "loss": 0.9735,
+      "step": 36
+    },
+    {
+      "epoch": 0.09588597343699384,
+      "grad_norm": 5.71875,
+      "learning_rate": 1.48e-05,
+      "loss": 0.9409,
+      "step": 37
+    },
+    {
+      "epoch": 0.09847748623258827,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.5200000000000002e-05,
+      "loss": 0.9108,
+      "step": 38
+    },
+    {
+      "epoch": 0.1010689990281827,
+      "grad_norm": 4.90625,
+      "learning_rate": 1.5600000000000003e-05,
+      "loss": 0.9029,
+      "step": 39
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "grad_norm": 5.0625,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.8938,
+      "step": 40
+    },
+    {
+      "epoch": 0.10366051182377713,
+      "eval_loss": 0.8692858815193176,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5016,
+      "eval_samples_per_second": 106.238,
+      "eval_steps_per_second": 13.305,
+      "step": 40
+    },
+    {
+      "epoch": 0.10625202461937155,
+      "grad_norm": 5.125,
+      "learning_rate": 1.64e-05,
+      "loss": 0.8501,
+      "step": 41
+    },
+    {
+      "epoch": 0.10884353741496598,
+      "grad_norm": 4.53125,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 0.8545,
+      "step": 42
+    },
+    {
+      "epoch": 0.11143505021056041,
+      "grad_norm": 5.1875,
+      "learning_rate": 1.72e-05,
+      "loss": 0.8169,
+      "step": 43
+    },
+    {
+      "epoch": 0.11402656300615484,
+      "grad_norm": 4.5,
+      "learning_rate": 1.76e-05,
+      "loss": 0.8102,
+      "step": 44
+    },
+    {
+      "epoch": 0.11661807580174927,
+      "grad_norm": 4.3125,
+      "learning_rate": 1.8e-05,
+      "loss": 0.8093,
+      "step": 45
+    },
+    {
+      "epoch": 0.1192095885973437,
+      "grad_norm": 4.28125,
+      "learning_rate": 1.8400000000000003e-05,
+      "loss": 0.7798,
+      "step": 46
+    },
+    {
+      "epoch": 0.12180110139293812,
+      "grad_norm": 4.09375,
+      "learning_rate": 1.88e-05,
+      "loss": 0.7761,
+      "step": 47
+    },
+    {
+      "epoch": 0.12439261418853255,
+      "grad_norm": 4.125,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.7647,
+      "step": 48
+    },
+    {
+      "epoch": 0.12698412698412698,
+      "grad_norm": 3.765625,
+      "learning_rate": 1.9600000000000002e-05,
+      "loss": 0.7612,
+      "step": 49
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "grad_norm": 4.0,
+      "learning_rate": 2e-05,
+      "loss": 0.7498,
+      "step": 50
+    },
+    {
+      "epoch": 0.12957563977972142,
+      "eval_loss": 0.744697093963623,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6181,
+      "eval_samples_per_second": 105.735,
+      "eval_steps_per_second": 13.242,
+      "step": 50
+    },
+    {
+      "epoch": 0.13216715257531583,
+      "grad_norm": 4.5,
+      "learning_rate": 1.9999560279257315e-05,
+      "loss": 0.7504,
+      "step": 51
+    },
+    {
+      "epoch": 0.13475866537091027,
+      "grad_norm": 3.796875,
+      "learning_rate": 1.999824115570012e-05,
+      "loss": 0.7181,
+      "step": 52
+    },
+    {
+      "epoch": 0.1373501781665047,
+      "grad_norm": 3.703125,
+      "learning_rate": 1.9996042745337617e-05,
+      "loss": 0.7195,
+      "step": 53
+    },
+    {
+      "epoch": 0.13994169096209913,
+      "grad_norm": 3.921875,
+      "learning_rate": 1.999296524150713e-05,
+      "loss": 0.7232,
+      "step": 54
+    },
+    {
+      "epoch": 0.14253320375769354,
+      "grad_norm": 3.6875,
+      "learning_rate": 1.9989008914857115e-05,
+      "loss": 0.725,
+      "step": 55
+    },
+    {
+      "epoch": 0.14512471655328799,
+      "grad_norm": 3.734375,
+      "learning_rate": 1.9984174113323352e-05,
+      "loss": 0.6972,
+      "step": 56
+    },
+    {
+      "epoch": 0.1477162293488824,
+      "grad_norm": 3.5,
+      "learning_rate": 1.9978461262098342e-05,
+      "loss": 0.6792,
+      "step": 57
+    },
+    {
+      "epoch": 0.15030774214447684,
+      "grad_norm": 4.0625,
+      "learning_rate": 1.9971870863593924e-05,
+      "loss": 0.7192,
+      "step": 58
+    },
+    {
+      "epoch": 0.15289925494007126,
+      "grad_norm": 3.75,
+      "learning_rate": 1.9964403497397086e-05,
+      "loss": 0.7107,
+      "step": 59
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9956059820218982e-05,
+      "loss": 0.6866,
+      "step": 60
+    },
+    {
+      "epoch": 0.1554907677356657,
+      "eval_loss": 0.6849498152732849,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5203,
+      "eval_samples_per_second": 106.157,
+      "eval_steps_per_second": 13.295,
+      "step": 60
+    },
+    {
+      "epoch": 0.1580822805312601,
+      "grad_norm": 3.953125,
+      "learning_rate": 1.9946840565837206e-05,
+      "loss": 0.6818,
+      "step": 61
+    },
+    {
+      "epoch": 0.16067379332685455,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.9936746545031225e-05,
+      "loss": 0.6855,
+      "step": 62
+    },
+    {
+      "epoch": 0.16326530612244897,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9925778645511112e-05,
+      "loss": 0.6799,
+      "step": 63
+    },
+    {
+      "epoch": 0.1658568189180434,
+      "grad_norm": 3.828125,
+      "learning_rate": 1.991393783183945e-05,
+      "loss": 0.6719,
+      "step": 64
+    },
+    {
+      "epoch": 0.16844833171363785,
+      "grad_norm": 3.0,
+      "learning_rate": 1.990122514534651e-05,
+      "loss": 0.66,
+      "step": 65
+    },
+    {
+      "epoch": 0.17103984450923226,
+      "grad_norm": 2.96875,
+      "learning_rate": 1.988764170403869e-05,
+      "loss": 0.6718,
+      "step": 66
+    },
+    {
+      "epoch": 0.1736313573048267,
+      "grad_norm": 3.15625,
+      "learning_rate": 1.9873188702500163e-05,
+      "loss": 0.6661,
+      "step": 67
+    },
+    {
+      "epoch": 0.17622287010042112,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.9857867411787847e-05,
+      "loss": 0.6523,
+      "step": 68
+    },
+    {
+      "epoch": 0.17881438289601556,
+      "grad_norm": 3.203125,
+      "learning_rate": 1.9841679179319605e-05,
+      "loss": 0.6494,
+      "step": 69
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "grad_norm": 2.875,
+      "learning_rate": 1.982462542875576e-05,
+      "loss": 0.6411,
+      "step": 70
+    },
+    {
+      "epoch": 0.18140589569160998,
+      "eval_loss": 0.6511064171791077,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5304,
+      "eval_samples_per_second": 106.113,
+      "eval_steps_per_second": 13.29,
+      "step": 70
+    },
+    {
+      "epoch": 0.18399740848720442,
+      "grad_norm": 3.296875,
+      "learning_rate": 1.9806707659873886e-05,
+      "loss": 0.6463,
+      "step": 71
+    },
+    {
+      "epoch": 0.18658892128279883,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9787927448436913e-05,
+      "loss": 0.6375,
+      "step": 72
+    },
+    {
+      "epoch": 0.18918043407839327,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9768286446054533e-05,
+      "loss": 0.6386,
+      "step": 73
+    },
+    {
+      "epoch": 0.1917719468739877,
+      "grad_norm": 2.875,
+      "learning_rate": 1.974778638003799e-05,
+      "loss": 0.6368,
+      "step": 74
+    },
+    {
+      "epoch": 0.19436345966958213,
+      "grad_norm": 2.953125,
+      "learning_rate": 1.972642905324813e-05,
+      "loss": 0.6271,
+      "step": 75
+    },
+    {
+      "epoch": 0.19695497246517654,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9704216343936875e-05,
+      "loss": 0.6501,
+      "step": 76
+    },
+    {
+      "epoch": 0.19954648526077098,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9681150205582025e-05,
+      "loss": 0.6111,
+      "step": 77
+    },
+    {
+      "epoch": 0.2021379980563654,
+      "grad_norm": 3.171875,
+      "learning_rate": 1.9657232666715485e-05,
+      "loss": 0.651,
+      "step": 78
+    },
+    {
+      "epoch": 0.20472951085195984,
+      "grad_norm": 3.0625,
+      "learning_rate": 1.9632465830744846e-05,
+      "loss": 0.6184,
+      "step": 79
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "grad_norm": 3.25,
+      "learning_rate": 1.9606851875768404e-05,
+      "loss": 0.6259,
+      "step": 80
+    },
+    {
+      "epoch": 0.20732102364755425,
+      "eval_loss": 0.6278263330459595,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4892,
+      "eval_samples_per_second": 106.292,
+      "eval_steps_per_second": 13.312,
+      "step": 80
+    },
+    {
+      "epoch": 0.2099125364431487,
+      "grad_norm": 2.875,
+      "learning_rate": 1.9580393054383622e-05,
+      "loss": 0.6288,
+      "step": 81
+    },
+    {
+      "epoch": 0.2125040492387431,
+      "grad_norm": 3.359375,
+      "learning_rate": 1.9553091693489018e-05,
+      "loss": 0.6403,
+      "step": 82
+    },
+    {
+      "epoch": 0.21509556203433755,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9524950194079534e-05,
+      "loss": 0.621,
+      "step": 83
+    },
+    {
+      "epoch": 0.21768707482993196,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9495971031035367e-05,
+      "loss": 0.6135,
+      "step": 84
+    },
+    {
+      "epoch": 0.2202785876255264,
+      "grad_norm": 2.921875,
+      "learning_rate": 1.9466156752904344e-05,
+      "loss": 0.6211,
+      "step": 85
+    },
+    {
+      "epoch": 0.22287010042112082,
+      "grad_norm": 3.109375,
+      "learning_rate": 1.9435509981677764e-05,
+      "loss": 0.6198,
+      "step": 86
+    },
+    {
+      "epoch": 0.22546161321671526,
+      "grad_norm": 2.625,
+      "learning_rate": 1.9404033412559828e-05,
+      "loss": 0.5972,
+      "step": 87
+    },
+    {
+      "epoch": 0.22805312601230968,
+      "grad_norm": 3.0,
+      "learning_rate": 1.9371729813730606e-05,
+      "loss": 0.614,
+      "step": 88
+    },
+    {
+      "epoch": 0.23064463880790412,
+      "grad_norm": 2.75,
+      "learning_rate": 1.9338602026102594e-05,
+      "loss": 0.5998,
+      "step": 89
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.9304652963070868e-05,
+      "loss": 0.6181,
+      "step": 90
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "eval_loss": 0.6108965277671814,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4982,
+      "eval_samples_per_second": 106.253,
+      "eval_steps_per_second": 13.307,
+      "step": 90
+    },
+    {
+      "epoch": 0.23582766439909297,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.9269885610256866e-05,
+      "loss": 0.6059,
+      "step": 91
+    },
+    {
+      "epoch": 0.2384191771946874,
+      "grad_norm": 2.734375,
+      "learning_rate": 1.9234303025245835e-05,
+      "loss": 0.6165,
+      "step": 92
+    },
+    {
+      "epoch": 0.24101068999028183,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.919790833731791e-05,
+      "loss": 0.6241,
+      "step": 93
+    },
+    {
+      "epoch": 0.24360220278587624,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.9160704747172935e-05,
+      "loss": 0.5913,
+      "step": 94
+    },
+    {
+      "epoch": 0.24619371558147068,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9122695526648968e-05,
+      "loss": 0.585,
+      "step": 95
+    },
+    {
+      "epoch": 0.2487852283770651,
+      "grad_norm": 3.484375,
+      "learning_rate": 1.9083884018434545e-05,
+      "loss": 0.5897,
+      "step": 96
+    },
+    {
+      "epoch": 0.2513767411726595,
+      "grad_norm": 2.84375,
+      "learning_rate": 1.9044273635774707e-05,
+      "loss": 0.6032,
+      "step": 97
+    },
+    {
+      "epoch": 0.25396825396825395,
+      "grad_norm": 2.75,
+      "learning_rate": 1.900386786217083e-05,
+      "loss": 0.594,
+      "step": 98
+    },
+    {
+      "epoch": 0.2565597667638484,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.8962670251074276e-05,
+      "loss": 0.5958,
+      "step": 99
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.8920684425573865e-05,
+      "loss": 0.5821,
+      "step": 100
+    },
+    {
+      "epoch": 0.25915127955944284,
+      "eval_loss": 0.5988509058952332,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4884,
+      "eval_samples_per_second": 106.295,
+      "eval_steps_per_second": 13.312,
+      "step": 100
+    },
+    {
+      "epoch": 0.2617427923550373,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.887791407807728e-05,
+      "loss": 0.5904,
+      "step": 101
+    },
+    {
+      "epoch": 0.26433430515063167,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.883436296998631e-05,
+      "loss": 0.5913,
+      "step": 102
+    },
+    {
+      "epoch": 0.2669258179462261,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.8790034931366073e-05,
+      "loss": 0.6114,
+      "step": 103
+    },
+    {
+      "epoch": 0.26951733074182055,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.8744933860608183e-05,
+      "loss": 0.5803,
+      "step": 104
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 3.40625,
+      "learning_rate": 1.8699063724087905e-05,
+      "loss": 0.5865,
+      "step": 105
+    },
+    {
+      "epoch": 0.2747003563330094,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.865242855581534e-05,
+      "loss": 0.586,
+      "step": 106
+    },
+    {
+      "epoch": 0.2772918691286038,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.8605032457080655e-05,
+      "loss": 0.5926,
+      "step": 107
+    },
+    {
+      "epoch": 0.27988338192419826,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.8556879596093393e-05,
+      "loss": 0.5927,
+      "step": 108
+    },
+    {
+      "epoch": 0.2824748947197927,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.850797420761592e-05,
+      "loss": 0.5937,
+      "step": 109
+    },
+    {
+      "epoch": 0.2850664075153871,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.8458320592590976e-05,
+      "loss": 0.5644,
+      "step": 110
+    },
+    {
+      "epoch": 0.2850664075153871,
+      "eval_loss": 0.5878391265869141,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.539,
+      "eval_samples_per_second": 106.076,
+      "eval_steps_per_second": 13.285,
+      "step": 110
+    },
+    {
+      "epoch": 0.28765792031098153,
+      "grad_norm": 2.75,
+      "learning_rate": 1.840792311776346e-05,
+      "loss": 0.5783,
+      "step": 111
+    },
+    {
+      "epoch": 0.29024943310657597,
+      "grad_norm": 2.375,
+      "learning_rate": 1.8356786215296387e-05,
+      "loss": 0.5813,
+      "step": 112
+    },
+    {
+      "epoch": 0.2928409459021704,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.83049143823811e-05,
+      "loss": 0.5788,
+      "step": 113
+    },
+    {
+      "epoch": 0.2954324586977648,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.825231218084178e-05,
+      "loss": 0.5831,
+      "step": 114
+    },
+    {
+      "epoch": 0.29802397149335924,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.8198984236734246e-05,
+      "loss": 0.5715,
+      "step": 115
+    },
+    {
+      "epoch": 0.3006154842889537,
+      "grad_norm": 2.546875,
+      "learning_rate": 1.8144935239939147e-05,
+      "loss": 0.5938,
+      "step": 116
+    },
+    {
+      "epoch": 0.3032069970845481,
+      "grad_norm": 2.90625,
+      "learning_rate": 1.8090169943749477e-05,
+      "loss": 0.5911,
+      "step": 117
+    },
+    {
+      "epoch": 0.3057985098801425,
+      "grad_norm": 2.859375,
+      "learning_rate": 1.8034693164452578e-05,
+      "loss": 0.5815,
+      "step": 118
+    },
+    {
+      "epoch": 0.30839002267573695,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.7978509780906578e-05,
+      "loss": 0.5804,
+      "step": 119
+    },
+    {
+      "epoch": 0.3109815354713314,
+      "grad_norm": 2.828125,
+      "learning_rate": 1.7921624734111292e-05,
+      "loss": 0.5749,
+      "step": 120
+    },
+    {
+      "epoch": 0.3109815354713314,
+      "eval_loss": 0.5794549584388733,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5897,
+      "eval_samples_per_second": 105.857,
+      "eval_steps_per_second": 13.258,
+      "step": 120
+    },
+    {
+      "epoch": 0.31357304826692584,
+      "grad_norm": 2.703125,
+      "learning_rate": 1.786404302677374e-05,
+      "loss": 0.5873,
+      "step": 121
+    },
+    {
+      "epoch": 0.3161645610625202,
+      "grad_norm": 2.375,
+      "learning_rate": 1.7805769722868132e-05,
+      "loss": 0.5856,
+      "step": 122
+    },
+    {
+      "epoch": 0.31875607385811466,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.7746809947190567e-05,
+      "loss": 0.578,
+      "step": 123
+    },
+    {
+      "epoch": 0.3213475866537091,
+      "grad_norm": 3.421875,
+      "learning_rate": 1.768716888490832e-05,
+      "loss": 0.5566,
+      "step": 124
+    },
+    {
+      "epoch": 0.32393909944930355,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.762685178110382e-05,
+      "loss": 0.5638,
+      "step": 125
+    },
+    {
+      "epoch": 0.32653061224489793,
+      "grad_norm": 3.125,
+      "learning_rate": 1.7565863940313416e-05,
+      "loss": 0.5827,
+      "step": 126
+    },
+    {
+      "epoch": 0.3291221250404924,
+      "grad_norm": 3.1875,
+      "learning_rate": 1.7504210726060828e-05,
+      "loss": 0.5596,
+      "step": 127
+    },
+    {
+      "epoch": 0.3317136378360868,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.7441897560385493e-05,
+      "loss": 0.5768,
+      "step": 128
+    },
+    {
+      "epoch": 0.33430515063168126,
+      "grad_norm": 2.546875,
+      "learning_rate": 1.7378929923365706e-05,
+      "loss": 0.5732,
+      "step": 129
+    },
+    {
+      "epoch": 0.3368966634272757,
+      "grad_norm": 2.78125,
+      "learning_rate": 1.731531335263669e-05,
+      "loss": 0.5781,
+      "step": 130
+    },
+    {
+      "epoch": 0.3368966634272757,
+      "eval_loss": 0.572753369808197,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5459,
+      "eval_samples_per_second": 106.046,
+      "eval_steps_per_second": 13.281,
+      "step": 130
+    },
+    {
+      "epoch": 0.3394881762228701,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.7251053442903597e-05,
+      "loss": 0.5689,
+      "step": 131
+    },
+    {
+      "epoch": 0.34207968901846453,
+      "grad_norm": 3.546875,
+      "learning_rate": 1.7186155845449465e-05,
+      "loss": 0.5695,
+      "step": 132
+    },
+    {
+      "epoch": 0.34467120181405897,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.712062626763825e-05,
+      "loss": 0.5627,
+      "step": 133
+    },
+    {
+      "epoch": 0.3472627146096534,
+      "grad_norm": 2.796875,
+      "learning_rate": 1.7054470472412876e-05,
+      "loss": 0.5759,
+      "step": 134
+    },
+    {
+      "epoch": 0.3498542274052478,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.698769427778842e-05,
+      "loss": 0.5662,
+      "step": 135
+    },
+    {
+      "epoch": 0.35244574020084224,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.6920303556340462e-05,
+      "loss": 0.5671,
+      "step": 136
+    },
+    {
+      "epoch": 0.3550372529964367,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.6852304234688627e-05,
+      "loss": 0.5572,
+      "step": 137
+    },
+    {
+      "epoch": 0.3576287657920311,
+      "grad_norm": 2.890625,
+      "learning_rate": 1.678370229297535e-05,
+      "loss": 0.5796,
+      "step": 138
+    },
+    {
+      "epoch": 0.3602202785876255,
+      "grad_norm": 3.03125,
+      "learning_rate": 1.6714503764339986e-05,
+      "loss": 0.5642,
+      "step": 139
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.664471473438822e-05,
+      "loss": 0.5711,
+      "step": 140
+    },
+    {
+      "epoch": 0.36281179138321995,
+      "eval_loss": 0.566669762134552,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5724,
+      "eval_samples_per_second": 105.932,
+      "eval_steps_per_second": 13.267,
+      "step": 140
+    },
+    {
+      "epoch": 0.3654033041788144,
+      "grad_norm": 2.625,
+      "learning_rate": 1.657434134065686e-05,
+      "loss": 0.5559,
+      "step": 141
+    },
+    {
+      "epoch": 0.36799481697440883,
+      "grad_norm": 3.015625,
+      "learning_rate": 1.6503389772074104e-05,
+      "loss": 0.575,
+      "step": 142
+    },
+    {
+      "epoch": 0.3705863297700032,
+      "grad_norm": 2.984375,
+      "learning_rate": 1.643186626841524e-05,
+      "loss": 0.5851,
+      "step": 143
+    },
+    {
+      "epoch": 0.37317784256559766,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.6359777119753887e-05,
+      "loss": 0.559,
+      "step": 144
+    },
+    {
+      "epoch": 0.3757693553611921,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.628712866590885e-05,
+      "loss": 0.5759,
+      "step": 145
+    },
+    {
+      "epoch": 0.37836086815678655,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.6213927295886547e-05,
+      "loss": 0.5579,
+      "step": 146
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.6140179447319134e-05,
+      "loss": 0.5717,
+      "step": 147
+    },
+    {
+      "epoch": 0.3835438937479754,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.6065891605898358e-05,
+      "loss": 0.5531,
+      "step": 148
+    },
+    {
+      "epoch": 0.3861354065435698,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.5991070304805184e-05,
+      "loss": 0.5358,
+      "step": 149
+    },
+    {
+      "epoch": 0.38872691933916426,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.5915722124135227e-05,
+      "loss": 0.5715,
+      "step": 150
+    },
+    {
+      "epoch": 0.38872691933916426,
+      "eval_loss": 0.5618191361427307,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5943,
+      "eval_samples_per_second": 105.838,
+      "eval_steps_per_second": 13.255,
+      "step": 150
+    },
+    {
+      "epoch": 0.39131843213475864,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.5839853690320074e-05,
+      "loss": 0.5413,
+      "step": 151
+    },
+    {
+      "epoch": 0.3939099449303531,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.5763471675544547e-05,
+      "loss": 0.5652,
+      "step": 152
+    },
+    {
+      "epoch": 0.3965014577259475,
+      "grad_norm": 2.609375,
+      "learning_rate": 1.5686582797159892e-05,
+      "loss": 0.5682,
+      "step": 153
+    },
+    {
+      "epoch": 0.39909297052154197,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.5609193817093057e-05,
+      "loss": 0.5577,
+      "step": 154
+    },
+    {
+      "epoch": 0.40168448331713635,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.5531311541251995e-05,
+      "loss": 0.5412,
+      "step": 155
+    },
+    {
+      "epoch": 0.4042759961127308,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.5452942818927144e-05,
+      "loss": 0.5441,
+      "step": 156
+    },
+    {
+      "epoch": 0.40686750890832524,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.5374094542189055e-05,
+      "loss": 0.5764,
+      "step": 157
+    },
+    {
+      "epoch": 0.4094590217039197,
+      "grad_norm": 3.078125,
+      "learning_rate": 1.52947736452823e-05,
+      "loss": 0.5763,
+      "step": 158
+    },
+    {
+      "epoch": 0.41205053449951407,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.521498710401561e-05,
+      "loss": 0.5482,
+      "step": 159
+    },
+    {
+      "epoch": 0.4146420472951085,
+      "grad_norm": 2.375,
+      "learning_rate": 1.513474193514842e-05,
+      "loss": 0.559,
+      "step": 160
+    },
+    {
+      "epoch": 0.4146420472951085,
+      "eval_loss": 0.5573883056640625,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5874,
+      "eval_samples_per_second": 105.867,
+      "eval_steps_per_second": 13.259,
+      "step": 160
+    },
+    {
+      "epoch": 0.41723356009070295,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.505404519577379e-05,
+      "loss": 0.5642,
+      "step": 161
+    },
+    {
+      "epoch": 0.4198250728862974,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.4972903982697744e-05,
+      "loss": 0.5535,
+      "step": 162
+    },
+    {
+      "epoch": 0.42241658568189183,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.4891325431815182e-05,
+      "loss": 0.5519,
+      "step": 163
+    },
+    {
+      "epoch": 0.4250080984774862,
+      "grad_norm": 2.0,
+      "learning_rate": 1.48093167174823e-05,
+      "loss": 0.5611,
+      "step": 164
+    },
+    {
+      "epoch": 0.42759961127308066,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.4726885051885654e-05,
+      "loss": 0.5556,
+      "step": 165
+    },
+    {
+      "epoch": 0.4301911240686751,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.4644037684407883e-05,
+      "loss": 0.5289,
+      "step": 166
+    },
+    {
+      "epoch": 0.43278263686426954,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.4560781900990187e-05,
+      "loss": 0.5538,
+      "step": 167
+    },
+    {
+      "epoch": 0.43537414965986393,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.4477125023491536e-05,
+      "loss": 0.5589,
+      "step": 168
+    },
+    {
+      "epoch": 0.43796566245545837,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.4393074409044802e-05,
+      "loss": 0.5423,
+      "step": 169
+    },
+    {
+      "epoch": 0.4405571752510528,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.4308637449409705e-05,
+      "loss": 0.5593,
+      "step": 170
+    },
+    {
+      "epoch": 0.4405571752510528,
+      "eval_loss": 0.553532063961029,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5188,
+      "eval_samples_per_second": 106.163,
+      "eval_steps_per_second": 13.296,
+      "step": 170
+    },
+    {
+      "epoch": 0.44314868804664725,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.4223821570322762e-05,
+      "loss": 0.5789,
+      "step": 171
+    },
+    {
+      "epoch": 0.44574020084224164,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.4138634230844242e-05,
+      "loss": 0.5693,
+      "step": 172
+    },
+    {
+      "epoch": 0.4483317136378361,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.4053082922702183e-05,
+      "loss": 0.5496,
+      "step": 173
+    },
+    {
+      "epoch": 0.4509232264334305,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.3967175169633537e-05,
+      "loss": 0.5362,
+      "step": 174
+    },
+    {
+      "epoch": 0.45351473922902497,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.3880918526722497e-05,
+      "loss": 0.5739,
+      "step": 175
+    },
+    {
+      "epoch": 0.45610625202461935,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.3794320579736084e-05,
+      "loss": 0.5493,
+      "step": 176
+    },
+    {
+      "epoch": 0.4586977648202138,
+      "grad_norm": 2.59375,
+      "learning_rate": 1.3707388944457007e-05,
+      "loss": 0.5665,
+      "step": 177
+    },
+    {
+      "epoch": 0.46128927761580824,
+      "grad_norm": 2.125,
+      "learning_rate": 1.3620131266013912e-05,
+      "loss": 0.5557,
+      "step": 178
+    },
+    {
+      "epoch": 0.4638807904114027,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.3532555218209037e-05,
+      "loss": 0.5402,
+      "step": 179
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "grad_norm": 2.375,
+      "learning_rate": 1.344466850284333e-05,
+      "loss": 0.5625,
+      "step": 180
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "eval_loss": 0.5502737164497375,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6233,
+      "eval_samples_per_second": 105.713,
+      "eval_steps_per_second": 13.239,
+      "step": 180
+    },
+    {
+      "epoch": 0.4690638160025915,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.335647884903915e-05,
+      "loss": 0.5492,
+      "step": 181
+    },
+    {
+      "epoch": 0.47165532879818595,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.3267994012560505e-05,
+      "loss": 0.5493,
+      "step": 182
+    },
+    {
+      "epoch": 0.4742468415937804,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.3179221775131005e-05,
+      "loss": 0.5508,
+      "step": 183
+    },
+    {
+      "epoch": 0.4768383543893748,
+      "grad_norm": 3.34375,
+      "learning_rate": 1.3090169943749475e-05,
+      "loss": 0.5427,
+      "step": 184
+    },
+    {
+      "epoch": 0.4794298671849692,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.300084635000341e-05,
+      "loss": 0.568,
+      "step": 185
+    },
+    {
+      "epoch": 0.48202137998056366,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.2911258849380199e-05,
+      "loss": 0.5478,
+      "step": 186
+    },
+    {
+      "epoch": 0.4846128927761581,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.2821415320576309e-05,
+      "loss": 0.5582,
+      "step": 187
+    },
+    {
+      "epoch": 0.4872044055717525,
+      "grad_norm": 3.0,
+      "learning_rate": 1.273132366480438e-05,
+      "loss": 0.5368,
+      "step": 188
+    },
+    {
+      "epoch": 0.4897959183673469,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.2640991805098369e-05,
+      "loss": 0.5608,
+      "step": 189
+    },
+    {
+      "epoch": 0.49238743116294137,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.2550427685616767e-05,
+      "loss": 0.5482,
+      "step": 190
+    },
+    {
+      "epoch": 0.49238743116294137,
+      "eval_loss": 0.547302782535553,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5429,
+      "eval_samples_per_second": 106.059,
+      "eval_steps_per_second": 13.283,
+      "step": 190
+    },
+    {
+      "epoch": 0.4949789439585358,
+      "grad_norm": 2.640625,
+      "learning_rate": 1.2459639270943944e-05,
+      "loss": 0.5508,
+      "step": 191
+    },
+    {
+      "epoch": 0.4975704567541302,
+      "grad_norm": 2.65625,
+      "learning_rate": 1.2368634545389732e-05,
+      "loss": 0.5402,
+      "step": 192
+    },
+    {
+      "epoch": 0.5001619695497247,
+      "grad_norm": 2.25,
+      "learning_rate": 1.2277421512287227e-05,
+      "loss": 0.5313,
+      "step": 193
+    },
+    {
+      "epoch": 0.502753482345319,
+      "grad_norm": 2.296875,
+      "learning_rate": 1.2186008193288963e-05,
+      "loss": 0.5535,
+      "step": 194
+    },
+    {
+      "epoch": 0.5053449951409135,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.2094402627661447e-05,
+      "loss": 0.5375,
+      "step": 195
+    },
+    {
+      "epoch": 0.5079365079365079,
+      "grad_norm": 2.375,
+      "learning_rate": 1.2002612871578143e-05,
+      "loss": 0.5564,
+      "step": 196
+    },
+    {
+      "epoch": 0.5105280207321024,
+      "grad_norm": 2.5,
+      "learning_rate": 1.1910646997411002e-05,
+      "loss": 0.5459,
+      "step": 197
+    },
+    {
+      "epoch": 0.5131195335276968,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.1818513093020514e-05,
+      "loss": 0.5545,
+      "step": 198
+    },
+    {
+      "epoch": 0.5157110463232912,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.172621926104446e-05,
+      "loss": 0.535,
+      "step": 199
+    },
+    {
+      "epoch": 0.5183025591188857,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.1633773618185302e-05,
+      "loss": 0.5469,
+      "step": 200
+    },
+    {
+      "epoch": 0.5183025591188857,
+      "eval_loss": 0.545219898223877,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.4828,
+      "eval_samples_per_second": 106.319,
+      "eval_steps_per_second": 13.315,
+      "step": 200
+    },
+    {
+      "epoch": 0.5208940719144801,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.1541184294496393e-05,
+      "loss": 0.5349,
+      "step": 201
+    },
+    {
+      "epoch": 0.5234855847100746,
+      "grad_norm": 2.46875,
+      "learning_rate": 1.144845943266696e-05,
+      "loss": 0.5271,
+      "step": 202
+    },
+    {
+      "epoch": 0.5260770975056689,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.1355607187306036e-05,
+      "loss": 0.5431,
+      "step": 203
+    },
+    {
+      "epoch": 0.5286686103012633,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.1262635724225273e-05,
+      "loss": 0.5344,
+      "step": 204
+    },
+    {
+      "epoch": 0.5312601230968578,
+      "grad_norm": 2.53125,
+      "learning_rate": 1.1169553219720828e-05,
+      "loss": 0.5386,
+      "step": 205
+    },
+    {
+      "epoch": 0.5338516358924522,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.1076367859854305e-05,
+      "loss": 0.5277,
+      "step": 206
+    },
+    {
+      "epoch": 0.5364431486880467,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.0983087839732834e-05,
+      "loss": 0.5339,
+      "step": 207
+    },
+    {
+      "epoch": 0.5390346614836411,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.0889721362788361e-05,
+      "loss": 0.5535,
+      "step": 208
+    },
+    {
+      "epoch": 0.5416261742792355,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.0796276640056201e-05,
+      "loss": 0.5385,
+      "step": 209
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.070276188945293e-05,
+      "loss": 0.5628,
+      "step": 210
+    },
+    {
+      "epoch": 0.54421768707483,
+      "eval_loss": 0.5431892275810242,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5521,
+      "eval_samples_per_second": 106.019,
+      "eval_steps_per_second": 13.278,
+      "step": 210
+    },
+    {
+      "epoch": 0.5468091998704243,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.060918533505367e-05,
+      "loss": 0.538,
+      "step": 211
+    },
+    {
+      "epoch": 0.5494007126660188,
+      "grad_norm": 3.1875,
+      "learning_rate": 1.0515555206368815e-05,
+      "loss": 0.53,
+      "step": 212
+    },
+    {
+      "epoch": 0.5519922254616132,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.0421879737620312e-05,
+      "loss": 0.5315,
+      "step": 213
+    },
+    {
+      "epoch": 0.5545837382572076,
+      "grad_norm": 2.125,
+      "learning_rate": 1.03281671670175e-05,
+      "loss": 0.5408,
+      "step": 214
+    },
+    {
+      "epoch": 0.5571752510528021,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.0234425736032607e-05,
+      "loss": 0.5436,
+      "step": 215
+    },
+    {
+      "epoch": 0.5597667638483965,
+      "grad_norm": 2.40625,
+      "learning_rate": 1.014066368867596e-05,
+      "loss": 0.5269,
+      "step": 216
+    },
+    {
+      "epoch": 0.562358276643991,
+      "grad_norm": 2.125,
+      "learning_rate": 1.0046889270770987e-05,
+      "loss": 0.5412,
+      "step": 217
+    },
+    {
+      "epoch": 0.5649497894395854,
+      "grad_norm": 2.3125,
+      "learning_rate": 9.953110729229018e-06,
+      "loss": 0.5346,
+      "step": 218
+    },
+    {
+      "epoch": 0.5675413022351797,
+      "grad_norm": 2.109375,
+      "learning_rate": 9.859336311324042e-06,
+      "loss": 0.5435,
+      "step": 219
+    },
+    {
+      "epoch": 0.5701328150307742,
+      "grad_norm": 2.125,
+      "learning_rate": 9.765574263967397e-06,
+      "loss": 0.5482,
+      "step": 220
+    },
+    {
+      "epoch": 0.5701328150307742,
+      "eval_loss": 0.5415154695510864,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.571,
+      "eval_samples_per_second": 105.938,
+      "eval_steps_per_second": 13.268,
+      "step": 220
+    },
+    {
+      "epoch": 0.5727243278263686,
+      "grad_norm": 2.21875,
+      "learning_rate": 9.671832832982502e-06,
+      "loss": 0.5401,
+      "step": 221
+    },
+    {
+      "epoch": 0.5753158406219631,
+      "grad_norm": 2.3125,
+      "learning_rate": 9.57812026237969e-06,
+      "loss": 0.5359,
+      "step": 222
+    },
+    {
+      "epoch": 0.5779073534175575,
+      "grad_norm": 2.375,
+      "learning_rate": 9.484444793631186e-06,
+      "loss": 0.5366,
+      "step": 223
+    },
+    {
+      "epoch": 0.5804988662131519,
+      "grad_norm": 2.09375,
+      "learning_rate": 9.390814664946332e-06,
+      "loss": 0.5343,
+      "step": 224
+    },
+    {
+      "epoch": 0.5830903790087464,
+      "grad_norm": 2.03125,
+      "learning_rate": 9.297238110547075e-06,
+      "loss": 0.5385,
+      "step": 225
+    },
+    {
+      "epoch": 0.5856818918043408,
+      "grad_norm": 2.21875,
+      "learning_rate": 9.203723359943802e-06,
+      "loss": 0.5473,
+      "step": 226
+    },
+    {
+      "epoch": 0.5882734045999352,
+      "grad_norm": 2.53125,
+      "learning_rate": 9.110278637211642e-06,
+      "loss": 0.5316,
+      "step": 227
+    },
+    {
+      "epoch": 0.5908649173955296,
+      "grad_norm": 2.046875,
+      "learning_rate": 9.016912160267168e-06,
+      "loss": 0.53,
+      "step": 228
+    },
+    {
+      "epoch": 0.593456430191124,
+      "grad_norm": 2.4375,
+      "learning_rate": 8.923632140145702e-06,
+      "loss": 0.5497,
+      "step": 229
+    },
+    {
+      "epoch": 0.5960479429867185,
+      "grad_norm": 2.15625,
+      "learning_rate": 8.830446780279175e-06,
+      "loss": 0.5499,
+      "step": 230
+    },
+    {
+      "epoch": 0.5960479429867185,
+      "eval_loss": 0.5400256514549255,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5551,
+      "eval_samples_per_second": 106.007,
+      "eval_steps_per_second": 13.276,
+      "step": 230
+    },
+    {
+      "epoch": 0.5986394557823129,
+      "grad_norm": 2.15625,
+      "learning_rate": 8.73736427577473e-06,
+      "loss": 0.5301,
+      "step": 231
+    },
+    {
+      "epoch": 0.6012309685779074,
+      "grad_norm": 2.171875,
+      "learning_rate": 8.644392812693969e-06,
+      "loss": 0.5542,
+      "step": 232
+    },
+    {
+      "epoch": 0.6038224813735018,
+      "grad_norm": 2.1875,
+      "learning_rate": 8.551540567333041e-06,
+      "loss": 0.5374,
+      "step": 233
+    },
+    {
+      "epoch": 0.6064139941690962,
+      "grad_norm": 2.421875,
+      "learning_rate": 8.458815705503612e-06,
+      "loss": 0.549,
+      "step": 234
+    },
+    {
+      "epoch": 0.6090055069646907,
+      "grad_norm": 2.0625,
+      "learning_rate": 8.366226381814698e-06,
+      "loss": 0.5383,
+      "step": 235
+    },
+    {
+      "epoch": 0.611597019760285,
+      "grad_norm": 2.203125,
+      "learning_rate": 8.273780738955545e-06,
+      "loss": 0.5297,
+      "step": 236
+    },
+    {
+      "epoch": 0.6141885325558795,
+      "grad_norm": 2.140625,
+      "learning_rate": 8.181486906979488e-06,
+      "loss": 0.5303,
+      "step": 237
+    },
+    {
+      "epoch": 0.6167800453514739,
+      "grad_norm": 2.109375,
+      "learning_rate": 8.089353002589003e-06,
+      "loss": 0.5429,
+      "step": 238
+    },
+    {
+      "epoch": 0.6193715581470683,
+      "grad_norm": 2.171875,
+      "learning_rate": 7.997387128421859e-06,
+      "loss": 0.5442,
+      "step": 239
+    },
+    {
+      "epoch": 0.6219630709426628,
+      "grad_norm": 2.328125,
+      "learning_rate": 7.905597372338558e-06,
+      "loss": 0.5451,
+      "step": 240
+    },
+    {
+      "epoch": 0.6219630709426628,
+      "eval_loss": 0.5389755964279175,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5608,
+      "eval_samples_per_second": 105.982,
+      "eval_steps_per_second": 13.273,
+      "step": 240
+    },
+    {
+      "epoch": 0.6245545837382572,
+      "grad_norm": 2.078125,
+      "learning_rate": 7.81399180671104e-06,
+      "loss": 0.5392,
+      "step": 241
+    },
+    {
+      "epoch": 0.6271460965338517,
+      "grad_norm": 2.140625,
+      "learning_rate": 7.722578487712776e-06,
+      "loss": 0.5303,
+      "step": 242
+    },
+    {
+      "epoch": 0.6297376093294461,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.631365454610273e-06,
+      "loss": 0.5387,
+      "step": 243
+    },
+    {
+      "epoch": 0.6323291221250404,
+      "grad_norm": 2.265625,
+      "learning_rate": 7.540360729056059e-06,
+      "loss": 0.5528,
+      "step": 244
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.449572314383237e-06,
+      "loss": 0.5488,
+      "step": 245
+    },
+    {
+      "epoch": 0.6375121477162293,
+      "grad_norm": 2.03125,
+      "learning_rate": 7.359008194901633e-06,
+      "loss": 0.5216,
+      "step": 246
+    },
+    {
+      "epoch": 0.6401036605118238,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.268676335195624e-06,
+      "loss": 0.5314,
+      "step": 247
+    },
+    {
+      "epoch": 0.6426951733074182,
+      "grad_norm": 2.125,
+      "learning_rate": 7.1785846794236945e-06,
+      "loss": 0.5411,
+      "step": 248
+    },
+    {
+      "epoch": 0.6452866861030127,
+      "grad_norm": 2.140625,
+      "learning_rate": 7.088741150619804e-06,
+      "loss": 0.5272,
+      "step": 249
+    },
+    {
+      "epoch": 0.6478781988986071,
+      "grad_norm": 2.125,
+      "learning_rate": 6.999153649996595e-06,
+      "loss": 0.5344,
+      "step": 250
+    },
+    {
+      "epoch": 0.6478781988986071,
+      "eval_loss": 0.538066029548645,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5108,
+      "eval_samples_per_second": 106.198,
+      "eval_steps_per_second": 13.3,
+      "step": 250
+    },
+    {
+      "epoch": 0.6504697116942015,
+      "grad_norm": 2.09375,
+      "learning_rate": 6.909830056250527e-06,
+      "loss": 0.532,
+      "step": 251
+    },
+    {
+      "epoch": 0.6530612244897959,
+      "grad_norm": 2.09375,
+      "learning_rate": 6.820778224868998e-06,
+      "loss": 0.5376,
+      "step": 252
+    },
+    {
+      "epoch": 0.6556527372853903,
+      "grad_norm": 2.15625,
+      "learning_rate": 6.732005987439494e-06,
+      "loss": 0.5501,
+      "step": 253
+    },
+    {
+      "epoch": 0.6582442500809847,
+      "grad_norm": 2.0625,
+      "learning_rate": 6.6435211509608546e-06,
+      "loss": 0.532,
+      "step": 254
+    },
+    {
+      "epoch": 0.6608357628765792,
+      "grad_norm": 2.3125,
+      "learning_rate": 6.555331497156671e-06,
+      "loss": 0.5406,
+      "step": 255
+    },
+    {
+      "epoch": 0.6634272756721736,
+      "grad_norm": 2.03125,
+      "learning_rate": 6.467444781790967e-06,
+      "loss": 0.5249,
+      "step": 256
+    },
+    {
+      "epoch": 0.6660187884677681,
+      "grad_norm": 2.171875,
+      "learning_rate": 6.379868733986089e-06,
+      "loss": 0.5195,
+      "step": 257
+    },
+    {
+      "epoch": 0.6686103012633625,
+      "grad_norm": 2.1875,
+      "learning_rate": 6.292611055542998e-06,
+      "loss": 0.5397,
+      "step": 258
+    },
+    {
+      "epoch": 0.671201814058957,
+      "grad_norm": 2.21875,
+      "learning_rate": 6.205679420263917e-06,
+      "loss": 0.5455,
+      "step": 259
+    },
+    {
+      "epoch": 0.6737933268545514,
+      "grad_norm": 2.09375,
+      "learning_rate": 6.119081473277502e-06,
+      "loss": 0.5414,
+      "step": 260
+    },
+    {
+      "epoch": 0.6737933268545514,
+      "eval_loss": 0.5372032523155212,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5567,
+      "eval_samples_per_second": 106.0,
+      "eval_steps_per_second": 13.275,
+      "step": 260
+    },
+    {
+      "epoch": 0.6763848396501457,
+      "grad_norm": 2.5,
+      "learning_rate": 6.032824830366466e-06,
+      "loss": 0.5398,
+      "step": 261
+    },
+    {
+      "epoch": 0.6789763524457402,
+      "grad_norm": 2.109375,
+      "learning_rate": 5.946917077297819e-06,
+      "loss": 0.5512,
+      "step": 262
+    },
+    {
+      "epoch": 0.6815678652413346,
+      "grad_norm": 2.125,
+      "learning_rate": 5.86136576915576e-06,
+      "loss": 0.5486,
+      "step": 263
+    },
+    {
+      "epoch": 0.6841593780369291,
+      "grad_norm": 2.203125,
+      "learning_rate": 5.7761784296772395e-06,
+      "loss": 0.543,
+      "step": 264
+    },
+    {
+      "epoch": 0.6867508908325235,
+      "grad_norm": 2.171875,
+      "learning_rate": 5.6913625505902966e-06,
+      "loss": 0.5065,
+      "step": 265
+    },
+    {
+      "epoch": 0.6893424036281179,
+      "grad_norm": 2.171875,
+      "learning_rate": 5.6069255909551986e-06,
+      "loss": 0.5328,
+      "step": 266
+    },
+    {
+      "epoch": 0.6919339164237124,
+      "grad_norm": 2.015625,
+      "learning_rate": 5.522874976508464e-06,
+      "loss": 0.5404,
+      "step": 267
+    },
+    {
+      "epoch": 0.6945254292193068,
+      "grad_norm": 1.9375,
+      "learning_rate": 5.439218099009822e-06,
+      "loss": 0.528,
+      "step": 268
+    },
+    {
+      "epoch": 0.6971169420149012,
+      "grad_norm": 2.0625,
+      "learning_rate": 5.355962315592118e-06,
+      "loss": 0.5377,
+      "step": 269
+    },
+    {
+      "epoch": 0.6997084548104956,
+      "grad_norm": 2.078125,
+      "learning_rate": 5.273114948114346e-06,
+      "loss": 0.5361,
+      "step": 270
+    },
+    {
+      "epoch": 0.6997084548104956,
+      "eval_loss": 0.5366343259811401,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5782,
+      "eval_samples_per_second": 105.907,
+      "eval_steps_per_second": 13.264,
+      "step": 270
+    },
+    {
+      "epoch": 0.70229996760609,
+      "grad_norm": 2.203125,
+      "learning_rate": 5.190683282517701e-06,
+      "loss": 0.5295,
+      "step": 271
+    },
+    {
+      "epoch": 0.7048914804016845,
+      "grad_norm": 2.15625,
+      "learning_rate": 5.108674568184822e-06,
+      "loss": 0.5333,
+      "step": 272
+    },
+    {
+      "epoch": 0.7074829931972789,
+      "grad_norm": 2.0,
+      "learning_rate": 5.027096017302261e-06,
+      "loss": 0.525,
+      "step": 273
+    },
+    {
+      "epoch": 0.7100745059928734,
+      "grad_norm": 2.140625,
+      "learning_rate": 4.945954804226214e-06,
+      "loss": 0.5328,
+      "step": 274
+    },
+    {
+      "epoch": 0.7126660187884678,
+      "grad_norm": 1.953125,
+      "learning_rate": 4.865258064851579e-06,
+      "loss": 0.5453,
+      "step": 275
+    },
+    {
+      "epoch": 0.7152575315840622,
+      "grad_norm": 2.109375,
+      "learning_rate": 4.785012895984397e-06,
+      "loss": 0.537,
+      "step": 276
+    },
+    {
+      "epoch": 0.7178490443796566,
+      "grad_norm": 1.9609375,
+      "learning_rate": 4.705226354717704e-06,
+      "loss": 0.5459,
+      "step": 277
+    },
+    {
+      "epoch": 0.720440557175251,
+      "grad_norm": 1.9296875,
+      "learning_rate": 4.625905457810943e-06,
+      "loss": 0.5444,
+      "step": 278
+    },
+    {
+      "epoch": 0.7230320699708455,
+      "grad_norm": 2.25,
+      "learning_rate": 4.547057181072861e-06,
+      "loss": 0.532,
+      "step": 279
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "grad_norm": 2.046875,
+      "learning_rate": 4.468688458748006e-06,
+      "loss": 0.5209,
+      "step": 280
+    },
+    {
+      "epoch": 0.7256235827664399,
+      "eval_loss": 0.5361185073852539,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5554,
+      "eval_samples_per_second": 106.005,
+      "eval_steps_per_second": 13.276,
+      "step": 280
+    },
+    {
+      "epoch": 0.7282150955620343,
+      "grad_norm": 1.9375,
+      "learning_rate": 4.390806182906946e-06,
+      "loss": 0.5304,
+      "step": 281
+    },
+    {
+      "epoch": 0.7308066083576288,
+      "grad_norm": 1.9921875,
+      "learning_rate": 4.313417202840106e-06,
+      "loss": 0.5532,
+      "step": 282
+    },
+    {
+      "epoch": 0.7333981211532232,
+      "grad_norm": 2.265625,
+      "learning_rate": 4.2365283244554545e-06,
+      "loss": 0.5358,
+      "step": 283
+    },
+    {
+      "epoch": 0.7359896339488177,
+      "grad_norm": 2.34375,
+      "learning_rate": 4.160146309679928e-06,
+      "loss": 0.5444,
+      "step": 284
+    },
+    {
+      "epoch": 0.738581146744412,
+      "grad_norm": 2.171875,
+      "learning_rate": 4.084277875864776e-06,
+      "loss": 0.5374,
+      "step": 285
+    },
+    {
+      "epoch": 0.7411726595400064,
+      "grad_norm": 2.53125,
+      "learning_rate": 4.008929695194819e-06,
+      "loss": 0.5441,
+      "step": 286
+    },
+    {
+      "epoch": 0.7437641723356009,
+      "grad_norm": 2.71875,
+      "learning_rate": 3.934108394101645e-06,
+      "loss": 0.5383,
+      "step": 287
+    },
+    {
+      "epoch": 0.7463556851311953,
+      "grad_norm": 2.3125,
+      "learning_rate": 3.859820552680867e-06,
+      "loss": 0.5481,
+      "step": 288
+    },
+    {
+      "epoch": 0.7489471979267898,
+      "grad_norm": 2.359375,
+      "learning_rate": 3.786072704113456e-06,
+      "loss": 0.5435,
+      "step": 289
+    },
+    {
+      "epoch": 0.7515387107223842,
+      "grad_norm": 2.0,
+      "learning_rate": 3.712871334091154e-06,
+      "loss": 0.5247,
+      "step": 290
+    },
+    {
+      "epoch": 0.7515387107223842,
+      "eval_loss": 0.5358829498291016,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5374,
+      "eval_samples_per_second": 106.083,
+      "eval_steps_per_second": 13.286,
+      "step": 290
+    },
+    {
+      "epoch": 0.7541302235179786,
+      "grad_norm": 2.03125,
+      "learning_rate": 3.640222880246117e-06,
+      "loss": 0.5215,
+      "step": 291
+    },
+    {
+      "epoch": 0.7567217363135731,
+      "grad_norm": 2.09375,
+      "learning_rate": 3.5681337315847677e-06,
+      "loss": 0.5337,
+      "step": 292
+    },
+    {
+      "epoch": 0.7593132491091675,
+      "grad_norm": 1.9609375,
+      "learning_rate": 3.4966102279258964e-06,
+      "loss": 0.5353,
+      "step": 293
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 1.984375,
+      "learning_rate": 3.425658659343141e-06,
+      "loss": 0.5454,
+      "step": 294
+    },
+    {
+      "epoch": 0.7644962747003563,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.355285265611784e-06,
+      "loss": 0.5172,
+      "step": 295
+    },
+    {
+      "epoch": 0.7670877874959507,
+      "grad_norm": 1.9765625,
+      "learning_rate": 3.285496235660013e-06,
+      "loss": 0.5267,
+      "step": 296
+    },
+    {
+      "epoch": 0.7696793002915452,
+      "grad_norm": 1.9765625,
+      "learning_rate": 3.216297707024655e-06,
+      "loss": 0.5436,
+      "step": 297
+    },
+    {
+      "epoch": 0.7722708130871396,
+      "grad_norm": 2.1875,
+      "learning_rate": 3.147695765311377e-06,
+      "loss": 0.5349,
+      "step": 298
+    },
+    {
+      "epoch": 0.7748623258827341,
+      "grad_norm": 2.0,
+      "learning_rate": 3.079696443659538e-06,
+      "loss": 0.5437,
+      "step": 299
+    },
+    {
+      "epoch": 0.7774538386783285,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.0123057222115835e-06,
+      "loss": 0.5185,
+      "step": 300
+    },
+    {
+      "epoch": 0.7774538386783285,
+      "eval_loss": 0.5356724858283997,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5736,
+      "eval_samples_per_second": 105.927,
+      "eval_steps_per_second": 13.266,
+      "step": 300
+    },
+    {
+      "epoch": 0.780045351473923,
+      "grad_norm": 2.109375,
+      "learning_rate": 2.94552952758713e-06,
+      "loss": 0.5204,
+      "step": 301
+    },
+    {
+      "epoch": 0.7826368642695173,
+      "grad_norm": 1.953125,
+      "learning_rate": 2.8793737323617554e-06,
+      "loss": 0.5483,
+      "step": 302
+    },
+    {
+      "epoch": 0.7852283770651117,
+      "grad_norm": 1.9140625,
+      "learning_rate": 2.813844154550537e-06,
+      "loss": 0.5433,
+      "step": 303
+    },
+    {
+      "epoch": 0.7878198898607062,
+      "grad_norm": 1.953125,
+      "learning_rate": 2.748946557096407e-06,
+      "loss": 0.5374,
+      "step": 304
+    },
+    {
+      "epoch": 0.7904114026563006,
+      "grad_norm": 2.125,
+      "learning_rate": 2.6846866473633126e-06,
+      "loss": 0.5405,
+      "step": 305
+    },
+    {
+      "epoch": 0.793002915451895,
+      "grad_norm": 2.0,
+      "learning_rate": 2.621070076634297e-06,
+      "loss": 0.535,
+      "step": 306
+    },
+    {
+      "epoch": 0.7955944282474895,
+      "grad_norm": 2.09375,
+      "learning_rate": 2.5581024396145116e-06,
+      "loss": 0.5474,
+      "step": 307
+    },
+    {
+      "epoch": 0.7981859410430839,
+      "grad_norm": 1.9296875,
+      "learning_rate": 2.4957892739391764e-06,
+      "loss": 0.5288,
+      "step": 308
+    },
+    {
+      "epoch": 0.8007774538386784,
+      "grad_norm": 2.453125,
+      "learning_rate": 2.434136059686587e-06,
+      "loss": 0.5437,
+      "step": 309
+    },
+    {
+      "epoch": 0.8033689666342727,
+      "grad_norm": 2.171875,
+      "learning_rate": 2.373148218896182e-06,
+      "loss": 0.5377,
+      "step": 310
+    },
+    {
+      "epoch": 0.8033689666342727,
+      "eval_loss": 0.5354843139648438,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.623,
+      "eval_samples_per_second": 105.714,
+      "eval_steps_per_second": 13.24,
+      "step": 310
+    },
+    {
+      "epoch": 0.8059604794298671,
+      "grad_norm": 2.09375,
+      "learning_rate": 2.312831115091683e-06,
+      "loss": 0.545,
+      "step": 311
+    },
+    {
+      "epoch": 0.8085519922254616,
+      "grad_norm": 1.984375,
+      "learning_rate": 2.2531900528094342e-06,
+      "loss": 0.5291,
+      "step": 312
+    },
+    {
+      "epoch": 0.811143505021056,
+      "grad_norm": 2.09375,
+      "learning_rate": 2.1942302771318713e-06,
+      "loss": 0.5306,
+      "step": 313
+    },
+    {
+      "epoch": 0.8137350178166505,
+      "grad_norm": 2.125,
+      "learning_rate": 2.135956973226262e-06,
+      "loss": 0.5352,
+      "step": 314
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 2.28125,
+      "learning_rate": 2.078375265888707e-06,
+      "loss": 0.5209,
+      "step": 315
+    },
+    {
+      "epoch": 0.8189180434078394,
+      "grad_norm": 2.109375,
+      "learning_rate": 2.021490219093426e-06,
+      "loss": 0.5331,
+      "step": 316
+    },
+    {
+      "epoch": 0.8215095562034338,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.9653068355474213e-06,
+      "loss": 0.5317,
+      "step": 317
+    },
+    {
+      "epoch": 0.8241010689990281,
+      "grad_norm": 2.171875,
+      "learning_rate": 1.9098300562505266e-06,
+      "loss": 0.5378,
+      "step": 318
+    },
+    {
+      "epoch": 0.8266925817946226,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.8550647600608573e-06,
+      "loss": 0.5243,
+      "step": 319
+    },
+    {
+      "epoch": 0.829284094590217,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.8010157632657544e-06,
+      "loss": 0.5229,
+      "step": 320
+    },
+    {
+      "epoch": 0.829284094590217,
+      "eval_loss": 0.5353964567184448,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5807,
+      "eval_samples_per_second": 105.896,
+      "eval_steps_per_second": 13.262,
+      "step": 320
+    },
+    {
+      "epoch": 0.8318756073858115,
+      "grad_norm": 2.21875,
+      "learning_rate": 1.7476878191582247e-06,
+      "loss": 0.5288,
+      "step": 321
+    },
+    {
+      "epoch": 0.8344671201814059,
+      "grad_norm": 1.90625,
+      "learning_rate": 1.6950856176189034e-06,
+      "loss": 0.5234,
+      "step": 322
+    },
+    {
+      "epoch": 0.8370586329770003,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.6432137847036144e-06,
+      "loss": 0.5598,
+      "step": 323
+    },
+    {
+      "epoch": 0.8396501457725948,
+      "grad_norm": 2.03125,
+      "learning_rate": 1.5920768822365418e-06,
+      "loss": 0.5302,
+      "step": 324
+    },
+    {
+      "epoch": 0.8422416585681892,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.5416794074090258e-06,
+      "loss": 0.5338,
+      "step": 325
+    },
+    {
+      "epoch": 0.8448331713637837,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.4920257923840865e-06,
+      "loss": 0.5408,
+      "step": 326
+    },
+    {
+      "epoch": 0.847424684159378,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.4431204039066082e-06,
+      "loss": 0.5297,
+      "step": 327
+    },
+    {
+      "epoch": 0.8500161969549724,
+      "grad_norm": 2.125,
+      "learning_rate": 1.3949675429193465e-06,
+      "loss": 0.518,
+      "step": 328
+    },
+    {
+      "epoch": 0.8526077097505669,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.347571444184661e-06,
+      "loss": 0.5271,
+      "step": 329
+    },
+    {
+      "epoch": 0.8551992225461613,
+      "grad_norm": 2.5625,
+      "learning_rate": 1.300936275912098e-06,
+      "loss": 0.5267,
+      "step": 330
+    },
+    {
+      "epoch": 0.8551992225461613,
+      "eval_loss": 0.5353734493255615,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.6222,
+      "eval_samples_per_second": 105.718,
+      "eval_steps_per_second": 13.24,
+      "step": 330
+    },
+    {
+      "epoch": 0.8577907353417558,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.2550661393918217e-06,
+      "loss": 0.5435,
+      "step": 331
+    },
+    {
+      "epoch": 0.8603822481373502,
+      "grad_norm": 2.25,
+      "learning_rate": 1.2099650686339304e-06,
+      "loss": 0.5395,
+      "step": 332
+    },
+    {
+      "epoch": 0.8629737609329446,
+      "grad_norm": 1.984375,
+      "learning_rate": 1.1656370300136944e-06,
+      "loss": 0.5279,
+      "step": 333
+    },
+    {
+      "epoch": 0.8655652737285391,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.122085921922723e-06,
+      "loss": 0.5208,
+      "step": 334
+    },
+    {
+      "epoch": 0.8681567865241334,
+      "grad_norm": 2.0625,
+      "learning_rate": 1.0793155744261352e-06,
+      "loss": 0.5387,
+      "step": 335
+    },
+    {
+      "epoch": 0.8707482993197279,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.0373297489257273e-06,
+      "loss": 0.557,
+      "step": 336
+    },
+    {
+      "epoch": 0.8733398121153223,
+      "grad_norm": 2.015625,
+      "learning_rate": 9.96132137829171e-07,
+      "loss": 0.5282,
+      "step": 337
+    },
+    {
+      "epoch": 0.8759313249109167,
+      "grad_norm": 2.109375,
+      "learning_rate": 9.557263642252946e-07,
+      "loss": 0.5393,
+      "step": 338
+    },
+    {
+      "epoch": 0.8785228377065112,
+      "grad_norm": 2.46875,
+      "learning_rate": 9.161159815654574e-07,
+      "loss": 0.5283,
+      "step": 339
+    },
+    {
+      "epoch": 0.8811143505021056,
+      "grad_norm": 2.265625,
+      "learning_rate": 8.773044733510338e-07,
+      "loss": 0.5238,
+      "step": 340
+    },
+    {
+      "epoch": 0.8811143505021056,
+      "eval_loss": 0.535314679145813,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5977,
+      "eval_samples_per_second": 105.823,
+      "eval_steps_per_second": 13.253,
+      "step": 340
+    },
+    {
+      "epoch": 0.8837058632977001,
+      "grad_norm": 1.9609375,
+      "learning_rate": 8.392952528270659e-07,
+      "loss": 0.5426,
+      "step": 341
+    },
+    {
+      "epoch": 0.8862973760932945,
+      "grad_norm": 3.515625,
+      "learning_rate": 8.020916626820919e-07,
+      "loss": 0.5267,
+      "step": 342
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 2.1875,
+      "learning_rate": 7.656969747541664e-07,
+      "loss": 0.5243,
+      "step": 343
+    },
+    {
+      "epoch": 0.8914804016844833,
+      "grad_norm": 2.0,
+      "learning_rate": 7.30114389743134e-07,
+      "loss": 0.5388,
+      "step": 344
+    },
+    {
+      "epoch": 0.8940719144800777,
+      "grad_norm": 1.9453125,
+      "learning_rate": 6.953470369291349e-07,
+      "loss": 0.547,
+      "step": 345
+    },
+    {
+      "epoch": 0.8966634272756722,
+      "grad_norm": 2.015625,
+      "learning_rate": 6.613979738974075e-07,
+      "loss": 0.5232,
+      "step": 346
+    },
+    {
+      "epoch": 0.8992549400712666,
+      "grad_norm": 2.109375,
+      "learning_rate": 6.282701862693963e-07,
+      "loss": 0.526,
+      "step": 347
+    },
+    {
+      "epoch": 0.901846452866861,
+      "grad_norm": 2.09375,
+      "learning_rate": 5.959665874401766e-07,
+      "loss": 0.5216,
+      "step": 348
+    },
+    {
+      "epoch": 0.9044379656624555,
+      "grad_norm": 2.875,
+      "learning_rate": 5.64490018322239e-07,
+      "loss": 0.5212,
+      "step": 349
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "grad_norm": 2.140625,
+      "learning_rate": 5.33843247095659e-07,
+      "loss": 0.5363,
+      "step": 350
+    },
+    {
+      "epoch": 0.9070294784580499,
+      "eval_loss": 0.5353750586509705,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5397,
+      "eval_samples_per_second": 106.073,
+      "eval_steps_per_second": 13.285,
+      "step": 350
+    },
+    {
+      "epoch": 0.9096209912536443,
+      "grad_norm": 2.1875,
+      "learning_rate": 5.040289689646338e-07,
+      "loss": 0.5363,
+      "step": 351
+    },
+    {
+      "epoch": 0.9122125040492387,
+      "grad_norm": 2.03125,
+      "learning_rate": 4.750498059204678e-07,
+      "loss": 0.5409,
+      "step": 352
+    },
+    {
+      "epoch": 0.9148040168448331,
+      "grad_norm": 2.015625,
+      "learning_rate": 4.469083065109825e-07,
+      "loss": 0.514,
+      "step": 353
+    },
+    {
+      "epoch": 0.9173955296404276,
+      "grad_norm": 2.21875,
+      "learning_rate": 4.196069456163787e-07,
+      "loss": 0.5272,
+      "step": 354
+    },
+    {
+      "epoch": 0.919987042436022,
+      "grad_norm": 1.859375,
+      "learning_rate": 3.931481242315993e-07,
+      "loss": 0.542,
+      "step": 355
+    },
+    {
+      "epoch": 0.9225785552316165,
+      "grad_norm": 1.96875,
+      "learning_rate": 3.6753416925515596e-07,
+      "loss": 0.5248,
+      "step": 356
+    },
+    {
+      "epoch": 0.9251700680272109,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.427673332845138e-07,
+      "loss": 0.5215,
+      "step": 357
+    },
+    {
+      "epoch": 0.9277615808228054,
+      "grad_norm": 1.90625,
+      "learning_rate": 3.188497944179758e-07,
+      "loss": 0.5354,
+      "step": 358
+    },
+    {
+      "epoch": 0.9303530936183998,
+      "grad_norm": 2.03125,
+      "learning_rate": 2.9578365606312666e-07,
+      "loss": 0.53,
+      "step": 359
+    },
+    {
+      "epoch": 0.9329446064139941,
+      "grad_norm": 2.109375,
+      "learning_rate": 2.735709467518699e-07,
+      "loss": 0.5356,
+      "step": 360
+    },
+    {
+      "epoch": 0.9329446064139941,
+      "eval_loss": 0.5352712273597717,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5757,
+      "eval_samples_per_second": 105.918,
+      "eval_steps_per_second": 13.265,
+      "step": 360
+    },
+    {
+      "epoch": 0.9355361192095886,
+      "grad_norm": 2.140625,
+      "learning_rate": 2.522136199620095e-07,
+      "loss": 0.5191,
+      "step": 361
+    },
+    {
+      "epoch": 0.938127632005183,
+      "grad_norm": 2.21875,
+      "learning_rate": 2.3171355394546625e-07,
+      "loss": 0.539,
+      "step": 362
+    },
+    {
+      "epoch": 0.9407191448007775,
+      "grad_norm": 2.0,
+      "learning_rate": 2.120725515630906e-07,
+      "loss": 0.5342,
+      "step": 363
+    },
+    {
+      "epoch": 0.9433106575963719,
+      "grad_norm": 2.203125,
+      "learning_rate": 1.932923401261133e-07,
+      "loss": 0.5382,
+      "step": 364
+    },
+    {
+      "epoch": 0.9459021703919663,
+      "grad_norm": 1.9296875,
+      "learning_rate": 1.7537457124423896e-07,
+      "loss": 0.5371,
+      "step": 365
+    },
+    {
+      "epoch": 0.9484936831875608,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.5832082068039545e-07,
+      "loss": 0.5303,
+      "step": 366
+    },
+    {
+      "epoch": 0.9510851959831552,
+      "grad_norm": 2.125,
+      "learning_rate": 1.421325882121538e-07,
+      "loss": 0.5178,
+      "step": 367
+    },
+    {
+      "epoch": 0.9536767087787495,
+      "grad_norm": 2.03125,
+      "learning_rate": 1.2681129749983811e-07,
+      "loss": 0.5309,
+      "step": 368
+    },
+    {
+      "epoch": 0.956268221574344,
+      "grad_norm": 2.09375,
+      "learning_rate": 1.1235829596131231e-07,
+      "loss": 0.536,
+      "step": 369
+    },
+    {
+      "epoch": 0.9588597343699384,
+      "grad_norm": 2.109375,
+      "learning_rate": 9.877485465349057e-08,
+      "loss": 0.5315,
+      "step": 370
+    },
+    {
+      "epoch": 0.9588597343699384,
+      "eval_loss": 0.5353068113327026,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5295,
+      "eval_samples_per_second": 106.117,
+      "eval_steps_per_second": 13.29,
+      "step": 370
+    },
+    {
+      "epoch": 0.9614512471655329,
+      "grad_norm": 1.9609375,
+      "learning_rate": 8.606216816055336e-08,
+      "loss": 0.5176,
+      "step": 371
+    },
+    {
+      "epoch": 0.9640427599611273,
+      "grad_norm": 2.03125,
+      "learning_rate": 7.422135448889034e-08,
+      "loss": 0.5252,
+      "step": 372
+    },
+    {
+      "epoch": 0.9666342727567218,
+      "grad_norm": 2.46875,
+      "learning_rate": 6.325345496877688e-08,
+      "loss": 0.5599,
+      "step": 373
+    },
+    {
+      "epoch": 0.9692257855523162,
+      "grad_norm": 2.140625,
+      "learning_rate": 5.315943416279834e-08,
+      "loss": 0.5342,
+      "step": 374
+    },
+    {
+      "epoch": 0.9718172983479106,
+      "grad_norm": 2.140625,
+      "learning_rate": 4.394017978101905e-08,
+      "loss": 0.5183,
+      "step": 375
+    },
+    {
+      "epoch": 0.974408811143505,
+      "grad_norm": 1.984375,
+      "learning_rate": 3.5596502602917024e-08,
+      "loss": 0.5256,
+      "step": 376
+    },
+    {
+      "epoch": 0.9770003239390994,
+      "grad_norm": 2.171875,
+      "learning_rate": 2.8129136406075397e-08,
+      "loss": 0.5406,
+      "step": 377
+    },
+    {
+      "epoch": 0.9795918367346939,
+      "grad_norm": 2.0,
+      "learning_rate": 2.1538737901658503e-08,
+      "loss": 0.5329,
+      "step": 378
+    },
+    {
+      "epoch": 0.9821833495302883,
+      "grad_norm": 1.9765625,
+      "learning_rate": 1.5825886676649195e-08,
+      "loss": 0.5575,
+      "step": 379
+    },
+    {
+      "epoch": 0.9847748623258827,
+      "grad_norm": 1.9375,
+      "learning_rate": 1.099108514288627e-08,
+      "loss": 0.5297,
+      "step": 380
+    },
+    {
+      "epoch": 0.9847748623258827,
+      "eval_loss": 0.535337507724762,
+      "eval_model_preparation_time": 0.0003,
+      "eval_runtime": 24.5539,
+      "eval_samples_per_second": 106.012,
+      "eval_steps_per_second": 13.277,
+      "step": 380
+    },
+    {
+      "epoch": 0.9873663751214772,
+      "grad_norm": 2.125,
+      "learning_rate": 7.034758492872051e-09,
+      "loss": 0.5427,
+      "step": 381
+    },
+    {
+      "epoch": 0.9899578879170716,
+      "grad_norm": 2.046875,
+      "learning_rate": 3.957254662385612e-09,
+      "loss": 0.558,
+      "step": 382
+    },
+    {
+      "epoch": 0.9925494007126661,
+      "grad_norm": 2.015625,
+      "learning_rate": 1.7588442998817124e-09,
+      "loss": 0.5265,
+      "step": 383
+    },
+    {
+      "epoch": 0.9951409135082604,
+      "grad_norm": 2.046875,
+      "learning_rate": 4.3972074268650023e-10,
+      "loss": 0.5434,
+      "step": 384
+    },
+    {
+      "epoch": 0.9977324263038548,
+      "grad_norm": 2.25,
+      "learning_rate": 0.0,
+      "loss": 0.5182,
+      "step": 385
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 385,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.071291410939904e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-385/training_args.bin b/checkpoint-385/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f7aa13c9cfd007e1e40a3fab731420d1b2409529
--- /dev/null
+++ b/checkpoint-385/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca15ebb16a59dc055f8ae01b1eb1038e869a6556aa3b7b802c58f2dde6a2993c
+size 5304
diff --git a/config.json b/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..4a3fd3ad7e0af9bc29478da576fa28f4642ce531
--- /dev/null
+++ b/config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322b2d3fc9de58487c02f1cd651d484d73244f9704de3d5feea36d002a38af0f
+size 748
diff --git a/generation_config.json b/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..475e140691313b73815f90f47ad5fde439467dfd
--- /dev/null
+++ b/generation_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82bdd024a0ba026e4001e0433d3922a27b1242913db7be87e2851d17165b99c8
+size 111
diff --git a/model.safetensors b/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..044e065a9d2f95c14465bea33ae152611e30f650
--- /dev/null
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:783f210de0e18c0cadd41c5bb8b4d1bbf18230cf2d1546ebf37450c870ceee8d
+size 810702192
diff --git a/special_tokens_map.json b/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..2a565ac6033dc9066dd29eadd1019f36f9709c86
--- /dev/null
+++ b/special_tokens_map.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57491904f8680d4b52ed440f1f7ba48cad1c31ecf3eb453b03484e6ff4723ae8
+size 473
diff --git a/tokenizer.json b/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..614d0562ad8d2db17f5aa36ec866a872f7d821a0
--- /dev/null
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:870f4e2baa6b683221fa52004d5d6f40ab8c9d31961617304b78c910c2c3caf2
+size 2114042
diff --git a/tokenizer_config.json b/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbc1d807e6afd67bc165abeb593b1c257efad4b3
--- /dev/null
+++ b/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4f587ac3a67c3964d3e84248f95ffc8e4f0e64d9de677692df39aa35eb2d11
+size 4845