diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-100/model.safetensors b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-100/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f524ff00e4286c80c34ed3d1408a7de698035d00
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-100/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8749be65766a222b429693237e64265c8cd1de9e5ecc640edd236a82c4104703
+size 326089504
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..de6beb590b0e8b3e9c8d04d3840a318e3ca32972
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b6b41b3992ab21b7e4314b4221e0216dc81143a31d7d952ce2026e5e9bbce7a
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..346bcb82e1636b039ae721df5cbf9f6e9b2f261c
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:705dc680cddb2763952d02bd0c4896df8f6c8ea9f23872a8803b9f7ff3212b22
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..73d5cebd041ca7da912777af79a2b4d79135c904
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ac2ca5ce745cc0887441aef0a4e6d64737f8e1e6bb181a9a0b6a086af876168
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4346e58cb81bd76bc78b303b9e4ab15d451fc019
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a238bc271c92ba0de016bcac42d6dc92985ab8c68bd1714524e869f9b87da77
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1700/model.safetensors b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1700/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..696ae2911dde5d95e86e1baf60c13b2d54378b31
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-1700/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdf1d86429f4ccda8e970d53fa8918a5ac37fe458e14f77253104b7add6997d6
+size 326089504
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5a22d1755dc2c80d1bd44e1211811a3c9abd5ee2
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ac4367f54d74a4483d4568e1ac859ca97fadc95fb1bc3026deddf88dc17ddab
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3a4d7c890b4d78b5cd56037b1005b4193b8e693e
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:072d48e62af09e7102e29a3af99d4caa718e186ed9b2ddd0718e3ee61325f326
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bfb0c4cd04245ffdf1d2f27247e2c05f3acbfc8a
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e59a404990ecc0b52d30b93a23edd1e2633bae1fc4b689a7b6795504bd3b8200
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b89bb32f5755165d744171b0fdb3ab5564cd77c6
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd2de2c47df439dbd7230ec705979f12d113ac759897563f5234820f561683b5
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-200/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..afc435bbfc9c300867b08b2a75934dbb391557b0
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ca9903723007bb50535944558afd060eebb34073ccec95f967b647aec48b2b3
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..587878b88e09bf3453929811d96880fdc0cc7bdc
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6453ac9b6f2973e71b1ac7342ce010ba6e9672b79c8f8bf63c1954921f3c3e7a
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..cce2425faf0cd2b67bbac68f9ed71f65ef648e63
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3dab17930e9127533cc897b1897c4c3ff9ed5d1a8fd7c9eafb1942f87d0d94a
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e1e02004797c00ee47f1fdc3c114a7b04e4a4473
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6341482dcda46a7bde41908e79ab6d4695fd45c96b278e9cbd379770af5a29bb
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c55616a558a238dd704411d1fb9deca8dbd25277
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b40bfe212186f14e11e59855fe9f5f8f5bed0f29bdb973b1dd4fb5415b1a7be
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a5f36ecf6e62e2ec5d8cd47402267a597a4c07ce
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5aaef4c6d2c1e078476ace23fe86ecbc8a814ed03b38152b3d6166f920b88c48
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..105db7520c64d8c9c881966c6524a3f4439dc981
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:420c4523b42fafbe8cf7a6c979b4a46825c218555e7dd7a3a0345f1c06e72447
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0365f9e4a6e14245ba3750490adad3a5634a6f71
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdf23f75398f66532ba69578a836725b2137d0c822eb32c05fae3b3ae4a15ffe
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2100/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..00f5098dc751758580d253c58dd16af7e9003fc4
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0742b74a7df60cb32c3a579f10edb2b1a40d6328d6aa9a30906fdd5b12916a89
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..72b4dafe96dda6f4c07c1312b93a8b0a3bb772b0
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e75b7622cb921bb4795b291325ddd72c6211bd74fa431b1c1650bd1dabf7adec
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3bae27a85d61f54a3256038f6ef2715bdf518d56
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:215c93dfcd83e61efaa1b05e96ed961a0f4bbb75cfa0020f84d4c4bcdcce0b65
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f9231ef1f11eb6c880ab6c9d4e0af05bdee4aba2
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c7b186683e21ff91e598e3446dfce11a325b04bf32b190af80f6980f17b3bb2
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9d6df447fd545e0032ec5282f056c7d9880e20b1
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aed3106d49821f58aafa535af384b055130c02083eab2de8d2d7a8ad5d624be5
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..24481f27f632a7f053a6c96334dd014713630e41
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca14835c889e06658caff479bd0e4660fded93fd288d87186545489f158c686a
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a9f2cf5980308ba92f6fd59227606c91b3e3766e
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9204708d3aafc75de9133d5cf7693adc0be6b74515defcb1e8bd6eae247ed91d
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0a386d00e49d825d95fc25f4715bd2bec04e923f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab40efd212baa6faa06da113355c3a1c7d67c1ecf0b874dc49116bdf850f1b0a
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..983f4b54a93def194c397da5b233d1566411091c
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c69bd10dccfa099cbe5cd2e585f6bb376fdbc77dfefc85351c92f27b1994286d
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5b84e79cbcbc19f18d2b7668d7cb7aa2cd11d66a
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63e8e35862c7ae335a3f8c87860d6e1ce49043aa9832ad2c2ac60dd246d8fa60
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..eb70c9cfb0c671f7c0c54e16137c1f7dc98a1b57
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b03c8bf083620d89d627b15b4db1159fc671bfa55b65243e052ccce269f18667
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..91fd3d477365b88230ce32331b6a57479e72c48f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6f3943d087eaef446a27ebcdd691c9bf0c3655ea600a0cca0187da4d59091bd
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7e66bca95268a50e046e1f6dfb576b71535894ba
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca37d65da813d6c9df55b3c08a4eaa88ed4d03aba1b61ce8d88fcb8c3215d1e9
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7820328d6471d757dbb6cbe87572a0b68ba57d9e
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:947cfe521ab51aa563ae3eb10f0f883d90c23d7d21c47f5d2cbf93e403884c5c
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e4bbe804f38ead8311b0fbdd7bd7ab9e6fa5168a
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9139e90c7ef5de08f336347c9793f6f88cb4f8a7b09063a07a91106777a82921
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7070b321ad3db257e1533aa1b36d009cf42de6e4
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9c2d595930ed023d8138b158e346093785c936e13dcd6c9bed13a487b0c60ae
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3ecba6805a968b3c71d283259a57b8c10f7cc925
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faea8304b502731eb21fb5d65be84220e7176a5b3d03a62d66fee9deb4f0005b
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..77331076bfd9016c7cf0acf906de83e259d03374
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d89669bf888838315e77a6dae29177631f0a9059d588c72f24633654db23d0f8
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6a2515b3f905f88afe6856c1fa4217ec88f90e3f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b75192282ecc801b320ca86994e848662296847898d6775a49278cf0899ee0cd
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..23be3383173af2ccaad5c2ef20fc51702d276258
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bb4d0795ea10b19056645ef8bf0547d1fee367b97c3182985d8a5dfc385aa3d
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6468e7f2cd519b9178f0627e0be9060ebe7572d9
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f523b589a3f0310142cdaf541d9edd194bc2e0964aa32533c56410df2781816
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a8c0142b4507e53818de079fd4b642dfecabbbe7
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb40363f61a4886d6564b29144cd513092b5c9d9052e87353372be81a9a658b0
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..89bfe4655dbe88a0cc84348c04cf0ff482c1bab3
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2775fd3e2a1a0da706330c847dafd1b7af0a3cc46bfdb28f211531c879aa8c3
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..20f9c84cb2b294ae519608d050d8e5cf5e84fe57
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d751bb5f16d642e6eff381b149aa6e77a9067db9b10fdb1a2158f7282988760
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2800/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..8c6a10469fc698acf51d68a09dd6bf3c2d64c130
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0bb936d26b59f7e6cddd7cd3a46b0354be5fbfbc4a0309f0d3200329ab72b59
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..198b470a8095d1ddfe932b804ce8a067d51a5061
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4332e324d3da1b3448e98cbe2e61fd2bace0e51900b4c6d8f6a30ec11b30486
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6fa35e5f606c5f0019c1877ebf3ab1bfaf8436c8
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03f9ffbc46ef7877fda3751a9ada6df77b3600ed4657053efb4207d5a6f407df
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..01d26fbc1f7a6457575ba20625307a69f02176dd
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19ebba8694d5e5cb06c1944a55189f428204ff5e53b4de463bf9b3a2b0d70b69
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-2900/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..8cc0dd4a7db0011cced5624990541ad9ba8581e9
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c4e8b48187078eefb7663d43c38eaa5773f4070941e9d6d400dc4813cc0765e
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..37fc0f8c8379e5c1ceb17f2e9242172e8fa357da
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c4367c3cece8b6f044ec2c272be98cd37dc50e9d558c788ec7a5031dbe95a58
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d0e51b698b85c65a1f45286f834810a920bcee85
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:820f72ea35dc0406deb96452e5b9776226736bf845017aefd77c24049260baa0
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..eb910ffaec0eaf3acd652d58dd35423a9aa8bdbf
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccf505e317fedc0cf8b110f5205ee3985762a6c32155d811925686d7f19b3704
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-300/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7c26e34d9f93a666bd936b613511d00af004f6b2
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c11ccbccfa990e3d808845420710884dba8c88cc6e826d5fd7ffa588bda6ee5c
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6ca9d29c81b999c35a40245404e49ba4e9166606
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bc2dc65c50ae12872db9cb418d43df0cf0df7e130a745aa9b581c6146c83d29
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bde77952686560bbbf884f452b85a21411c9d480
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c599b61156fd00c02c2382bf843960524e5d2e9783793c6376e42186cf552d7
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c8930ea4cb7a95d0ae79cf69dc0dc8c6404f8cfb
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:985f23a71b306667a5d4ab6562968db84233838303e70f4ed165767567813f1f
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-700/training_args.bin b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-700/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3887a0e52e72e32427921c286630421cdb2d7885
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-700/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:882e399105a571c4656ff1fe3caeaa9bb79b911caa1367b83e187114714d9505
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-900/model.safetensors b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-900/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1f82f7482193dbf616d8bc96db049680c0758a2a
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_control_10M_seed0/runs/checkpoint-900/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23907b3dcd1e11df2c4893aec144b3080675cfb467bf3ac90907aef3233df849
+size 326089504
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f337ddf5f9b948eb7d548ab07f2bb22e44d88e18
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fd02406d4ae3161d068c376954f25db14b1f37d1a196a9517d6b5ec50ac169c
+size 497765391
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..23e7610030316d01c964d388d4e51b7e47f3b49c
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:632ea7ffb7772bba009228d75e53eca8403b45a2c4a902b880faa83e147afe2d
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..acbcd7ad5a76c5916a910b71403952627229e8a6
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-1000/global_step1000/zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:562646af93a72ba68ff20e54e008a4fe4c2751381ee5c234290c314fea67c0fc
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2100/global_step2100/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2100/global_step2100/zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c960591cf9388ebc86085c13484cd383b87fb792
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2100/global_step2100/zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:699260522cb2ef4122e3dd02ce1bd1a06aa5eb3274cf7b1f0cf0d98f4338ceae
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2100/global_step2100/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2100/global_step2100/zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c40bd97fa1190c5c4c2821d14bb49a27c290e026
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2100/global_step2100/zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45db0d05a73c5c8db9496a93b748e8561631707687161497a886a88d9c5f3229
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2200/global_step2200/mp_rank_00_model_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2200/global_step2200/mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..43173c19ca518a1b2bd38ac2d45721580775133f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2200/global_step2200/mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee3e07d1084dfd4d6e2ab3399729896ff39b78ca93df1de6d3e1e1d2fdea6b4b
+size 248917159
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2200/model.safetensors b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2200/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4a4e7868743011d0743d87103072d363ead91c96
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2200/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f31a9b35c74166195f7b519f6690c32e6bc6f43a6a86806819a3d27c2921150a
+size 326089504
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d7304d4299d712074271aa5efe94ca0a5d37590d
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb0fad9ad95527749368656664910012d97bc3cd58625b5bf8fa9286d87bbee9
+size 497765391
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e1ed552f3cc1c0299a8fb895542658aabd318e31
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1cba06e2728baca2aa9a7bdd3499223c227f7b76c408a6773bc8880a252fa31
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d98c46c295c380893f59349b7db26cbb46d2b203
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2300/global_step2300/zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d9337e7641213ce15948c6b40290e8a629bb1be91cfe27df9130e02831d1c6c
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c118934b4280da87a887ea8c6b93625e55e0a53
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d46770a0c18c8abb74ed9ee8d48b7fc87b0a9eadeffe413f4a3bf4dbf2bd6a6
+size 497765391
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ddff314e780df02f4ea48898cdc395743437400d
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7c66cd4fb74695ddf5700484ed9ec1d9e90962b765de1005ce931b41e355280
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..68b22a2dbfb002596e02d698c1b887023d5d2ad4
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2400/global_step2400/zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6cde08c7ad24936788c6e3b66c3c9e54936ac6d7e5f02ecfa6340b05a97da21
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2500/global_step2500/mp_rank_00_model_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2500/global_step2500/mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c1dc6616d5fd8fca407bd7d91935f7bd3f78bb73
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2500/global_step2500/mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4697b0dfcd793401cbc06aeab37d689c4df111672630d60a8ee5ed27325b22fd
+size 248917159
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2500/model.safetensors b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2500/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..76ce646f4eb0fc48e4b2beacd778527ff1d0146e
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2500/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4eba10aecb2ec4232b86b62f5f67b687999bfbd1f834a9195be22fb44528d319
+size 326089504
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2600/model.safetensors b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2600/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..efd435e6a81dd167ba62cf94f2d544fcdd038b93
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2600/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ead15e443ed49b0c99e76b2112ebcce1893f73c649efe2d480017486931669d
+size 326089504
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c15697804b17d01c3219accd033518c91998c2f6
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7405fcb0d2773db75421ef439d9760e8f073f2cdafc4fb6dae5953938255f597
+size 497765391
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e2c9166a2c16e88c106395e03230172fa3c7e351
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62258f00d53c56204ffd27d381facb35336cf4813a551d9a1026acb8481494ff
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a66316dbdd797e796186a7e757ee0bb7920264e1
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-2900/global_step2900/zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce179f8f79893fc2237733f05c530f7de31b189868d498e6f76941fa7abbb505
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6f4b9cdf782598c567069dd0cd5e10c7c5b52140
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bd007256867639ce21fee1f5494567cce6c08d567daebd646666a1c17e4a253
+size 497765391
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..700eb63b1a7c2949e0230ca94b56b9497d833bc6
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b55c6f7a1e8d9a01681bd8107e757b96455a47920ea067e513e84333800c6d15
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ebab89571c8e65b59c5ab45a9ea9071b57afbf12
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_full_10M_seed0/runs/checkpoint-300/global_step300/zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b47e9a861bac694bbaeb201faf40325b6bdba92947b10d094aa0e8f51e461fe9
+size 497766927
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9517baa4714cbea054b9cf2886e03ecf12e76cec
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51c4b9b0cb4121f8147348543c3a76e45ed25b30cc736f35e909f3aee5f44436
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..556ca58b7fb41c9b525c7ce311af945c47c82374
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2535b743eea20f2766691f1796fbaafa2bda2bc54125f3880a2ddc01027243f
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3fedcb3ff92fcdbbd76fe20937e3dd2eca94cf81
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f83976de2d9261fb2fe7946ebd689abd67bcf08f5fda8c65ac5781042870be1d
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..220d8e1780738ed164388392e14c0303ef18704d
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc2ab44a0fd4825bad2072a003107eb46e81d141fb685512fc1860db8b23f387
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-100/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..335c90b7b5d6db9869d01b998bc5da8b7e4b3c97
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6997d01b752f15e109c27651b1230033c310ef5d4f39f2753edca7e5754b4c4f
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1abfbf97e11dd77fe3a56395774f56ae9ca03c1c
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0992b694703ae98fb181fbc7aa9c38794b3f4cf06d27987a48b54e38c52183a2
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2c471caecc4020bfb7a589bab0b3361162353ec9
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:170bc158a2d2fe81582771b5720edef3f3ef60e1ec06f3cd6ebf3cc2c08e39a4
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1b683d92def26bf9b6ec06978beb5740110fe3eb
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8c897b993648152e2289ec96f470f849760f3b8bf4ac6270c8afa967c478578
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1100/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..904cb7a56c06fbaea384cb0dd05f541700a2b60f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:514ee00cecf5a440bc000c4917dd878b369a87b84239091df5fb505e0ab8fceb
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2d16d3524fbc5ce93778725f02506c1fa68c601a
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb87a0782672a4ba60ac9fd6985aadaf2de2b312a48ff2ae79f58e102295adc4
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..00bdce578bf61d80eb9783e5696502cb9a7799a8
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ac884e22840da938f1bc9697b22bc0e40405f1b453ba6c7c12d89a0d88fcb66
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..eceb04f40140c95c276868463eb9957f91e98b6c
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6813367728a28ef8ef6c9574cf3f57808010ff2c9c9d8e906e26753f1b544710
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1200/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..56083cf2bc307472aaa7f01595941ce5526705a0
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54d05f3f08088a9a8d2bc885e529871ca592b9d499743545ecdb7fb459d48872
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..417db73b802cd58631a4b625ef3d99de7c25509b
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17fbbc4082e74783f2730409f060152fd0fefb58fc40f7baaf9ccc93394ac66a
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1e405107c9ae92c2aacd3301c1c155841acda673
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06b8b0b7fc5e4d8172177f37db8cc5d48cf03e0c63c2e75e6acb73c6fe09a9a6
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0024ae65d14e1c950c8b1ebe72e26f0aebec69bb
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6048e178d7ba3c3808212eeb906a931437cbb1242dab0c6d3108fab2096f6bbb
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1300/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..894a4f92e06ae96fdc0c13ff9312a7a6feeaa865
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7e99302fcf427602d4777289ed1f4cedd77d903aa4bf67593fe02218d5c4a43
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..358b4b9bf7ad0f45f4701ca0472e422992d11b8f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b519b793b1316502eaacc42c09124663c80399152ce8435bd67f154c93d7798
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..cb564ceebb67722becd3cb6a82919113f203fd10
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:154140e471a24ed41a2e225cf3d38b2900a47269e7bd1e713e6b2a3d6e24a7b5
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b4b867ba03ac52ad59e5b233689549260f8d1909
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e59fd03c1e193fcac43fad01f7dd88ced49bc0b17daf56e24417283e3414186
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f595b4ee4112b2b9c55d976791ced9723a407217
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b5c6622530a1a9b6b2bf96c8bd5e51380053678ec1035c6699958ff8e709eb4
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4297fe04b6c3ec27a677e117b7e956769953e749
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:131295ac1517d2d23ff7ec06e4f02fa8158bf4a7f54b48a3875f346382d4cbad
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5e0dd987c317aea8ba11cb18fbc27a3505abff25
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b587e60906f004691b92c3aaede7dc71ec9518c020f84ec2d059668121018082
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7e48c5c7fbc5b9933ea0273edd7fa9f05f758e4c
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:998206324b9f0c329412a3c1ec2bff2bc6a878913170cbffe418e1f15728b43e
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1600/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6e0fe879aaa086507071f25996bcac37c9edd0eb
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4da1463b0823ccb1dcfbe22835431c0689fae24241d4bc68f8a44bd38691e95
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b93f23560555c071760b67cae286e839f5b07b99
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:621c9e33245fe4855f26944f680db4426c6f62489eef948bc83de4ee97bd20b8
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..abcb3eabc3821991a0606d914332cbaf1661106d
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3a61bf4d4643d5e98ff0eebe9f4bbb9528dfe4b1730a3169d0b364ac773904c
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3af7317c5d266b3b0d74783311ca6e4707f8b381
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93fe71b94f41a14e135a8c2e8f1ce9d74ab6384d40e2b261aac1c4124bbc3987
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-1700/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..76da3c23ec9979135d54c9a71afc604acc9c46ed
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4bf46106979defbbcd15302bd8835d516e576b0f38be79da9f4b422b56bf92e
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ff0f29613f7ebca138412cb840ad0d26ce1b09f1
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8058aeedf05963bd7d0956b4076b48970c075b326764707cd9cbc26465d653f
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2021ad2ef4c132abbc910dfd1e402f4e99478811
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9db8efb490f5423266f7e945490405c7b95b367efbbc92b8e5fac731b1d6b759
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c590b5bbd9b9a67cb0bbb8650542446fd9cbfe92
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59ba07e07713a48dd3185609e73eb0f8e693d26a2c54268cc97a248392c417e8
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-2700/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c932df69a412e735eedcb853ef26d53b93abac10
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0e23bf6a7152b2f60a2bb190a0aa0831fea91fa4edb0fac8ceb8e28386a3f2e
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..11c404ee19b35abc689a33468df4066dc600ff7f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59c36c7796a739fb2dece1d445d975dec4cab6dad3f76c133e29db70668009b9
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b2655604d6dd16a8c31ba6946fb80b74562daa87
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-400/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f681505801aa0fad5d4ac9bc5a7f3aa7953620c6739d00617acfefc119aefeb8
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6d82c64cad661269583f11b5d51ac2cfe0cf240b
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcd424d4994defe3292f934d62d1ae81be36f4c909aefa7a470d8a9d3ee15863
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_1.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f8f646a40f0acd9327a7d1228182cc029e4bd70e
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1b5d82b3468dd8dcd4b3b74ebbbda4f7b0ccb2acbeaeb11d35df2d69a70069
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f813a44f6ca56eba45bff5c44a25c480becb6393
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b399af941be374a1d41f8ae8b19d2b8ee3737d2684269545be9a45662716f850
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a7ced2d40f684a7f14c5dfc346ef418ad3f5c7dc
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96ef46628c4ea19d56138217b6f3750c6870b0ba796e47543e9d2b71c45060e2
+size 627
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/training_args.bin b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b9c6310f232ce991a14691700385427742356e43
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-600/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb0a6f24460a35a5849475791340c8d7dfb54225c5749b791b10e833071abfa
+size 6011
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/rng_state_0.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7498f004223e01dc677a891756621469082f8082
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:080cab4baedef52a2d6ebe5b6d1c0ffa1263a0ff0c15887e914b06a43116e15f
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/rng_state_2.pth b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..00835f5b69908e7f2ec8b32f6bd21723625aee7f
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0f9ca99b349a61e4d36c90867c6be2926f762286a9b02af6581b766cf0ac039
+size 16567
diff --git a/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/scheduler.pt b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f82a3a79310ff17a2cf892abfd2e1e8eb1ed7b91
--- /dev/null
+++ b/checkpoints/GPT-2/babylm_reverse_partial_10M_seed0/runs/checkpoint-900/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8db8e99a3c9acbe9a3218aec503d607b2de5c0aa0aa411686c1306c4065e255
+size 627
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/latest b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/latest
new file mode 100644
index 0000000000000000000000000000000000000000..d71c4c79bfb8e74842ee005fbfc27bedb903c39b
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/latest
@@ -0,0 +1 @@
+global_step1300
\ No newline at end of file
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/tokenizer_config.json b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b782d4602b361f7577abb52b710a8c0e088588d
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/tokenizer_config.json
@@ -0,0 +1,2078 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|finetune_right_pad_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|python_tag|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128256": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128257": {
+      "content": "🅁",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|end_of_text|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "[PAD]",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/zero_to_fp32.py b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/zero_to_fp32.py
new file mode 100644
index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1300/zero_to_fp32.py
@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/config.json b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/config.json
@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "meta-llama/Llama-3.2-3B",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 24,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float16",
+  "transformers_version": "4.45.1",
+  "use_cache": true,
+  "vocab_size": 128256
+}
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/generation_config.json b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/generation_config.json
@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": 128001,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.45.1"
+}
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/model.safetensors.index.json b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/model.safetensors.index.json
new file mode 100644
index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/model.safetensors.index.json
@@ -0,0 +1,262 @@
+{
+  "metadata": {
+    "total_size": 7213504512
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/special_tokens_map.json b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/special_tokens_map.json
@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/trainer_state.json b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b39d0ccf7f44726167300bd9cacafe7522aedd5d
--- /dev/null
+++ b/checkpoints/Llama-3.2-3B-FTP/babylm_reverse_full_10M_seed0/runs/checkpoint-1500/trainer_state.json
@@ -0,0 +1,11733 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.5408320493066254,
+  "eval_steps": 10,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.001027221366204417,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7525,
+      "step": 1
+    },
+    {
+      "epoch": 0.002054442732408834,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7605,
+      "step": 2
+    },
+    {
+      "epoch": 0.0030816640986132513,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.8008,
+      "step": 3
+    },
+    {
+      "epoch": 0.004108885464817668,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.809,
+      "step": 4
+    },
+    {
+      "epoch": 0.005136106831022085,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7841,
+      "step": 5
+    },
+    {
+      "epoch": 0.0061633281972265025,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7924,
+      "step": 6
+    },
+    {
+      "epoch": 0.007190549563430919,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7732,
+      "step": 7
+    },
+    {
+      "epoch": 0.008217770929635337,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7953,
+      "step": 8
+    },
+    {
+      "epoch": 0.009244992295839754,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7747,
+      "step": 9
+    },
+    {
+      "epoch": 0.01027221366204417,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7909,
+      "step": 10
+    },
+    {
+      "epoch": 0.01027221366204417,
+      "eval_loss": 2.813098192214966,
+      "eval_runtime": 38.1187,
+      "eval_samples_per_second": 26.234,
+      "eval_steps_per_second": 1.102,
+      "step": 10
+    },
+    {
+      "epoch": 0.011299435028248588,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7592,
+      "step": 11
+    },
+    {
+      "epoch": 0.012326656394453005,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7394,
+      "step": 12
+    },
+    {
+      "epoch": 0.01335387776065742,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.8111,
+      "step": 13
+    },
+    {
+      "epoch": 0.014381099126861838,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 2.7786,
+      "step": 14
+    },
+    {
+      "epoch": 0.015408320493066256,
+      "grad_norm": 5.814355850219727,
+      "learning_rate": 1.7123287671232876e-08,
+      "loss": 2.7711,
+      "step": 15
+    },
+    {
+      "epoch": 0.016435541859270673,
+      "grad_norm": 5.814355850219727,
+      "learning_rate": 1.7123287671232876e-08,
+      "loss": 2.7622,
+      "step": 16
+    },
+    {
+      "epoch": 0.01746276322547509,
+      "grad_norm": 5.814355850219727,
+      "learning_rate": 1.7123287671232876e-08,
+      "loss": 2.7291,
+      "step": 17
+    },
+    {
+      "epoch": 0.01848998459167951,
+      "grad_norm": 6.8249125480651855,
+      "learning_rate": 3.424657534246575e-08,
+      "loss": 2.7995,
+      "step": 18
+    },
+    {
+      "epoch": 0.019517205957883924,
+      "grad_norm": 5.581949710845947,
+      "learning_rate": 5.136986301369863e-08,
+      "loss": 2.5998,
+      "step": 19
+    },
+    {
+      "epoch": 0.02054442732408834,
+      "grad_norm": 3.0888617038726807,
+      "learning_rate": 6.84931506849315e-08,
+      "loss": 2.6031,
+      "step": 20
+    },
+    {
+      "epoch": 0.02054442732408834,
+      "eval_loss": 2.60642147064209,
+      "eval_runtime": 38.2226,
+      "eval_samples_per_second": 26.163,
+      "eval_steps_per_second": 1.099,
+      "step": 20
+    },
+    {
+      "epoch": 0.02157164869029276,
+      "grad_norm": 4.42941427230835,
+      "learning_rate": 8.561643835616439e-08,
+      "loss": 2.5816,
+      "step": 21
+    },
+    {
+      "epoch": 0.022598870056497175,
+      "grad_norm": 2.359813690185547,
+      "learning_rate": 1.0273972602739726e-07,
+      "loss": 2.5352,
+      "step": 22
+    },
+    {
+      "epoch": 0.02362609142270159,
+      "grad_norm": 2.9022927284240723,
+      "learning_rate": 1.1986301369863014e-07,
+      "loss": 2.4641,
+      "step": 23
+    },
+    {
+      "epoch": 0.02465331278890601,
+      "grad_norm": 1.6895997524261475,
+      "learning_rate": 1.36986301369863e-07,
+      "loss": 2.4582,
+      "step": 24
+    },
+    {
+      "epoch": 0.025680534155110426,
+      "grad_norm": 2.4555776119232178,
+      "learning_rate": 1.541095890410959e-07,
+      "loss": 2.4171,
+      "step": 25
+    },
+    {
+      "epoch": 0.02670775552131484,
+      "grad_norm": 1.9735779762268066,
+      "learning_rate": 1.7123287671232878e-07,
+      "loss": 2.4249,
+      "step": 26
+    },
+    {
+      "epoch": 0.02773497688751926,
+      "grad_norm": 2.71161150932312,
+      "learning_rate": 1.8835616438356165e-07,
+      "loss": 2.3791,
+      "step": 27
+    },
+    {
+      "epoch": 0.028762198253723677,
+      "grad_norm": 1.8482733964920044,
+      "learning_rate": 2.0547945205479452e-07,
+      "loss": 2.374,
+      "step": 28
+    },
+    {
+      "epoch": 0.029789419619928096,
+      "grad_norm": 2.191246747970581,
+      "learning_rate": 2.226027397260274e-07,
+      "loss": 2.3557,
+      "step": 29
+    },
+    {
+      "epoch": 0.030816640986132512,
+      "grad_norm": 2.0575971603393555,
+      "learning_rate": 2.397260273972603e-07,
+      "loss": 2.3315,
+      "step": 30
+    },
+    {
+      "epoch": 0.030816640986132512,
+      "eval_loss": 2.399639844894409,
+      "eval_runtime": 37.737,
+      "eval_samples_per_second": 26.499,
+      "eval_steps_per_second": 1.113,
+      "step": 30
+    },
+    {
+      "epoch": 0.03184386235233693,
+      "grad_norm": 2.0346758365631104,
+      "learning_rate": 2.568493150684932e-07,
+      "loss": 2.2791,
+      "step": 31
+    },
+    {
+      "epoch": 0.03287108371854135,
+      "grad_norm": 1.8710602521896362,
+      "learning_rate": 2.73972602739726e-07,
+      "loss": 2.3219,
+      "step": 32
+    },
+    {
+      "epoch": 0.03389830508474576,
+      "grad_norm": 1.9982675313949585,
+      "learning_rate": 2.910958904109589e-07,
+      "loss": 2.273,
+      "step": 33
+    },
+    {
+      "epoch": 0.03492552645095018,
+      "grad_norm": 2.092076301574707,
+      "learning_rate": 3.082191780821918e-07,
+      "loss": 2.2555,
+      "step": 34
+    },
+    {
+      "epoch": 0.035952747817154594,
+      "grad_norm": 1.870373249053955,
+      "learning_rate": 3.2534246575342466e-07,
+      "loss": 2.2314,
+      "step": 35
+    },
+    {
+      "epoch": 0.03697996918335902,
+      "grad_norm": 1.8378052711486816,
+      "learning_rate": 3.4246575342465755e-07,
+      "loss": 2.1244,
+      "step": 36
+    },
+    {
+      "epoch": 0.03800719054956343,
+      "grad_norm": 2.0156054496765137,
+      "learning_rate": 3.595890410958904e-07,
+      "loss": 2.2502,
+      "step": 37
+    },
+    {
+      "epoch": 0.03903441191576785,
+      "grad_norm": 1.6175312995910645,
+      "learning_rate": 3.767123287671233e-07,
+      "loss": 2.1566,
+      "step": 38
+    },
+    {
+      "epoch": 0.040061633281972264,
+      "grad_norm": 1.9051450490951538,
+      "learning_rate": 3.938356164383562e-07,
+      "loss": 2.167,
+      "step": 39
+    },
+    {
+      "epoch": 0.04108885464817668,
+      "grad_norm": 2.383063793182373,
+      "learning_rate": 4.1095890410958903e-07,
+      "loss": 2.2099,
+      "step": 40
+    },
+    {
+      "epoch": 0.04108885464817668,
+      "eval_loss": 2.2763354778289795,
+      "eval_runtime": 37.8085,
+      "eval_samples_per_second": 26.449,
+      "eval_steps_per_second": 1.111,
+      "step": 40
+    },
+    {
+      "epoch": 0.042116076014381096,
+      "grad_norm": 2.1310720443725586,
+      "learning_rate": 4.2808219178082193e-07,
+      "loss": 2.1714,
+      "step": 41
+    },
+    {
+      "epoch": 0.04314329738058552,
+      "grad_norm": 1.771941900253296,
+      "learning_rate": 4.452054794520548e-07,
+      "loss": 2.0895,
+      "step": 42
+    },
+    {
+      "epoch": 0.044170518746789934,
+      "grad_norm": 2.3784172534942627,
+      "learning_rate": 4.6232876712328767e-07,
+      "loss": 2.1555,
+      "step": 43
+    },
+    {
+      "epoch": 0.04519774011299435,
+      "grad_norm": 1.6692813634872437,
+      "learning_rate": 4.794520547945206e-07,
+      "loss": 2.11,
+      "step": 44
+    },
+    {
+      "epoch": 0.046224961479198766,
+      "grad_norm": 2.1780643463134766,
+      "learning_rate": 4.965753424657534e-07,
+      "loss": 2.1256,
+      "step": 45
+    },
+    {
+      "epoch": 0.04725218284540318,
+      "grad_norm": 1.9390568733215332,
+      "learning_rate": 5.136986301369864e-07,
+      "loss": 2.0971,
+      "step": 46
+    },
+    {
+      "epoch": 0.048279404211607604,
+      "grad_norm": 2.276216506958008,
+      "learning_rate": 5.308219178082192e-07,
+      "loss": 2.1479,
+      "step": 47
+    },
+    {
+      "epoch": 0.04930662557781202,
+      "grad_norm": 2.148573160171509,
+      "learning_rate": 5.47945205479452e-07,
+      "loss": 2.0808,
+      "step": 48
+    },
+    {
+      "epoch": 0.050333846944016436,
+      "grad_norm": 2.259422540664673,
+      "learning_rate": 5.65068493150685e-07,
+      "loss": 2.0823,
+      "step": 49
+    },
+    {
+      "epoch": 0.05136106831022085,
+      "grad_norm": 1.7703359127044678,
+      "learning_rate": 5.821917808219178e-07,
+      "loss": 2.0236,
+      "step": 50
+    },
+    {
+      "epoch": 0.05136106831022085,
+      "eval_loss": 2.2037274837493896,
+      "eval_runtime": 37.8385,
+      "eval_samples_per_second": 26.428,
+      "eval_steps_per_second": 1.11,
+      "step": 50
+    },
+    {
+      "epoch": 0.05238828967642527,
+      "grad_norm": 2.56268310546875,
+      "learning_rate": 5.993150684931507e-07,
+      "loss": 2.0228,
+      "step": 51
+    },
+    {
+      "epoch": 0.05341551104262968,
+      "grad_norm": 2.5256848335266113,
+      "learning_rate": 6.164383561643836e-07,
+      "loss": 2.1401,
+      "step": 52
+    },
+    {
+      "epoch": 0.054442732408834106,
+      "grad_norm": 2.6329638957977295,
+      "learning_rate": 6.335616438356165e-07,
+      "loss": 2.0107,
+      "step": 53
+    },
+    {
+      "epoch": 0.05546995377503852,
+      "grad_norm": 2.477762222290039,
+      "learning_rate": 6.506849315068493e-07,
+      "loss": 2.0648,
+      "step": 54
+    },
+    {
+      "epoch": 0.05649717514124294,
+      "grad_norm": 1.8869198560714722,
+      "learning_rate": 6.678082191780823e-07,
+      "loss": 2.0214,
+      "step": 55
+    },
+    {
+      "epoch": 0.05752439650744735,
+      "grad_norm": 2.707345485687256,
+      "learning_rate": 6.849315068493151e-07,
+      "loss": 1.9893,
+      "step": 56
+    },
+    {
+      "epoch": 0.05855161787365177,
+      "grad_norm": 2.067230701446533,
+      "learning_rate": 7.020547945205481e-07,
+      "loss": 2.0308,
+      "step": 57
+    },
+    {
+      "epoch": 0.05957883923985619,
+      "grad_norm": 2.1961982250213623,
+      "learning_rate": 7.191780821917808e-07,
+      "loss": 1.948,
+      "step": 58
+    },
+    {
+      "epoch": 0.06060606060606061,
+      "grad_norm": 2.1032280921936035,
+      "learning_rate": 7.363013698630137e-07,
+      "loss": 2.0098,
+      "step": 59
+    },
+    {
+      "epoch": 0.061633281972265024,
+      "grad_norm": 2.5191144943237305,
+      "learning_rate": 7.534246575342466e-07,
+      "loss": 2.0183,
+      "step": 60
+    },
+    {
+      "epoch": 0.061633281972265024,
+      "eval_loss": 2.1474344730377197,
+      "eval_runtime": 37.8631,
+      "eval_samples_per_second": 26.411,
+      "eval_steps_per_second": 1.109,
+      "step": 60
+    },
+    {
+      "epoch": 0.06266050333846944,
+      "grad_norm": 2.177739143371582,
+      "learning_rate": 7.705479452054795e-07,
+      "loss": 1.9702,
+      "step": 61
+    },
+    {
+      "epoch": 0.06368772470467386,
+      "grad_norm": 2.499784231185913,
+      "learning_rate": 7.876712328767124e-07,
+      "loss": 1.9825,
+      "step": 62
+    },
+    {
+      "epoch": 0.06471494607087827,
+      "grad_norm": 2.181591749191284,
+      "learning_rate": 8.047945205479453e-07,
+      "loss": 2.0414,
+      "step": 63
+    },
+    {
+      "epoch": 0.0657421674370827,
+      "grad_norm": 1.9575999975204468,
+      "learning_rate": 8.219178082191781e-07,
+      "loss": 2.0115,
+      "step": 64
+    },
+    {
+      "epoch": 0.0667693888032871,
+      "grad_norm": 2.3431336879730225,
+      "learning_rate": 8.39041095890411e-07,
+      "loss": 1.8781,
+      "step": 65
+    },
+    {
+      "epoch": 0.06779661016949153,
+      "grad_norm": 2.042747974395752,
+      "learning_rate": 8.561643835616439e-07,
+      "loss": 1.9798,
+      "step": 66
+    },
+    {
+      "epoch": 0.06882383153569595,
+      "grad_norm": 2.784104585647583,
+      "learning_rate": 8.732876712328768e-07,
+      "loss": 1.9274,
+      "step": 67
+    },
+    {
+      "epoch": 0.06985105290190036,
+      "grad_norm": 1.9781445264816284,
+      "learning_rate": 8.904109589041097e-07,
+      "loss": 2.0036,
+      "step": 68
+    },
+    {
+      "epoch": 0.07087827426810478,
+      "grad_norm": 2.631537437438965,
+      "learning_rate": 9.075342465753426e-07,
+      "loss": 1.938,
+      "step": 69
+    },
+    {
+      "epoch": 0.07190549563430919,
+      "grad_norm": 1.902701497077942,
+      "learning_rate": 9.246575342465753e-07,
+      "loss": 2.0545,
+      "step": 70
+    },
+    {
+      "epoch": 0.07190549563430919,
+      "eval_loss": 2.115269422531128,
+      "eval_runtime": 37.887,
+      "eval_samples_per_second": 26.394,
+      "eval_steps_per_second": 1.109,
+      "step": 70
+    },
+    {
+      "epoch": 0.07293271700051361,
+      "grad_norm": 1.9323904514312744,
+      "learning_rate": 9.417808219178083e-07,
+      "loss": 1.9463,
+      "step": 71
+    },
+    {
+      "epoch": 0.07395993836671803,
+      "grad_norm": 2.1631956100463867,
+      "learning_rate": 9.589041095890411e-07,
+      "loss": 1.8982,
+      "step": 72
+    },
+    {
+      "epoch": 0.07498715973292244,
+      "grad_norm": 1.959518313407898,
+      "learning_rate": 9.76027397260274e-07,
+      "loss": 1.9538,
+      "step": 73
+    },
+    {
+      "epoch": 0.07601438109912687,
+      "grad_norm": 1.9834569692611694,
+      "learning_rate": 9.931506849315068e-07,
+      "loss": 1.9257,
+      "step": 74
+    },
+    {
+      "epoch": 0.07704160246533127,
+      "grad_norm": 2.2818732261657715,
+      "learning_rate": 1.0102739726027399e-06,
+      "loss": 1.9318,
+      "step": 75
+    },
+    {
+      "epoch": 0.0780688238315357,
+      "grad_norm": 1.7696406841278076,
+      "learning_rate": 1.0273972602739727e-06,
+      "loss": 1.912,
+      "step": 76
+    },
+    {
+      "epoch": 0.07909604519774012,
+      "grad_norm": 2.396303176879883,
+      "learning_rate": 1.0445205479452056e-06,
+      "loss": 1.9482,
+      "step": 77
+    },
+    {
+      "epoch": 0.08012326656394453,
+      "grad_norm": 2.2371881008148193,
+      "learning_rate": 1.0616438356164384e-06,
+      "loss": 1.9448,
+      "step": 78
+    },
+    {
+      "epoch": 0.08115048793014895,
+      "grad_norm": 2.40077543258667,
+      "learning_rate": 1.0787671232876712e-06,
+      "loss": 1.8764,
+      "step": 79
+    },
+    {
+      "epoch": 0.08217770929635336,
+      "grad_norm": 2.267916679382324,
+      "learning_rate": 1.095890410958904e-06,
+      "loss": 1.9478,
+      "step": 80
+    },
+    {
+      "epoch": 0.08217770929635336,
+      "eval_loss": 2.0851893424987793,
+      "eval_runtime": 37.9176,
+      "eval_samples_per_second": 26.373,
+      "eval_steps_per_second": 1.108,
+      "step": 80
+    },
+    {
+      "epoch": 0.08320493066255778,
+      "grad_norm": 2.286259651184082,
+      "learning_rate": 1.1130136986301371e-06,
+      "loss": 1.8856,
+      "step": 81
+    },
+    {
+      "epoch": 0.08423215202876219,
+      "grad_norm": 2.160848379135132,
+      "learning_rate": 1.13013698630137e-06,
+      "loss": 1.9196,
+      "step": 82
+    },
+    {
+      "epoch": 0.08525937339496661,
+      "grad_norm": 2.3618931770324707,
+      "learning_rate": 1.1472602739726028e-06,
+      "loss": 1.9229,
+      "step": 83
+    },
+    {
+      "epoch": 0.08628659476117104,
+      "grad_norm": 2.1963961124420166,
+      "learning_rate": 1.1643835616438357e-06,
+      "loss": 1.8998,
+      "step": 84
+    },
+    {
+      "epoch": 0.08731381612737545,
+      "grad_norm": 2.197981357574463,
+      "learning_rate": 1.1815068493150685e-06,
+      "loss": 1.8716,
+      "step": 85
+    },
+    {
+      "epoch": 0.08834103749357987,
+      "grad_norm": 2.3562612533569336,
+      "learning_rate": 1.1986301369863014e-06,
+      "loss": 1.9336,
+      "step": 86
+    },
+    {
+      "epoch": 0.08936825885978428,
+      "grad_norm": 2.49712872505188,
+      "learning_rate": 1.2157534246575344e-06,
+      "loss": 1.8827,
+      "step": 87
+    },
+    {
+      "epoch": 0.0903954802259887,
+      "grad_norm": 2.315244436264038,
+      "learning_rate": 1.2328767123287673e-06,
+      "loss": 1.8801,
+      "step": 88
+    },
+    {
+      "epoch": 0.09142270159219312,
+      "grad_norm": 2.774587869644165,
+      "learning_rate": 1.25e-06,
+      "loss": 1.8972,
+      "step": 89
+    },
+    {
+      "epoch": 0.09244992295839753,
+      "grad_norm": 2.262364387512207,
+      "learning_rate": 1.267123287671233e-06,
+      "loss": 1.8426,
+      "step": 90
+    },
+    {
+      "epoch": 0.09244992295839753,
+      "eval_loss": 2.069462299346924,
+      "eval_runtime": 37.9319,
+      "eval_samples_per_second": 26.363,
+      "eval_steps_per_second": 1.107,
+      "step": 90
+    },
+    {
+      "epoch": 0.09347714432460195,
+      "grad_norm": 2.6195597648620605,
+      "learning_rate": 1.284246575342466e-06,
+      "loss": 1.871,
+      "step": 91
+    },
+    {
+      "epoch": 0.09450436569080636,
+      "grad_norm": 2.051847219467163,
+      "learning_rate": 1.3013698630136986e-06,
+      "loss": 1.8922,
+      "step": 92
+    },
+    {
+      "epoch": 0.09553158705701079,
+      "grad_norm": 2.5930886268615723,
+      "learning_rate": 1.3184931506849317e-06,
+      "loss": 1.8818,
+      "step": 93
+    },
+    {
+      "epoch": 0.09655880842321521,
+      "grad_norm": 2.304164171218872,
+      "learning_rate": 1.3356164383561645e-06,
+      "loss": 1.8779,
+      "step": 94
+    },
+    {
+      "epoch": 0.09758602978941962,
+      "grad_norm": 2.187354564666748,
+      "learning_rate": 1.3527397260273976e-06,
+      "loss": 1.9272,
+      "step": 95
+    },
+    {
+      "epoch": 0.09861325115562404,
+      "grad_norm": 2.2492458820343018,
+      "learning_rate": 1.3698630136986302e-06,
+      "loss": 1.805,
+      "step": 96
+    },
+    {
+      "epoch": 0.09964047252182845,
+      "grad_norm": 2.3325352668762207,
+      "learning_rate": 1.386986301369863e-06,
+      "loss": 1.8817,
+      "step": 97
+    },
+    {
+      "epoch": 0.10066769388803287,
+      "grad_norm": 2.1708617210388184,
+      "learning_rate": 1.4041095890410961e-06,
+      "loss": 1.7956,
+      "step": 98
+    },
+    {
+      "epoch": 0.1016949152542373,
+      "grad_norm": 2.4267706871032715,
+      "learning_rate": 1.421232876712329e-06,
+      "loss": 1.8116,
+      "step": 99
+    },
+    {
+      "epoch": 0.1027221366204417,
+      "grad_norm": 2.3286805152893066,
+      "learning_rate": 1.4383561643835616e-06,
+      "loss": 1.8299,
+      "step": 100
+    },
+    {
+      "epoch": 0.1027221366204417,
+      "eval_loss": 2.0549135208129883,
+      "eval_runtime": 37.935,
+      "eval_samples_per_second": 26.361,
+      "eval_steps_per_second": 1.107,
+      "step": 100
+    },
+    {
+      "epoch": 0.10374935798664613,
+      "grad_norm": 2.5391643047332764,
+      "learning_rate": 1.4554794520547946e-06,
+      "loss": 1.8521,
+      "step": 101
+    },
+    {
+      "epoch": 0.10477657935285054,
+      "grad_norm": 2.018380880355835,
+      "learning_rate": 1.4726027397260275e-06,
+      "loss": 1.8209,
+      "step": 102
+    },
+    {
+      "epoch": 0.10580380071905496,
+      "grad_norm": 2.3706746101379395,
+      "learning_rate": 1.4897260273972605e-06,
+      "loss": 1.79,
+      "step": 103
+    },
+    {
+      "epoch": 0.10683102208525937,
+      "grad_norm": 2.1290764808654785,
+      "learning_rate": 1.5068493150684932e-06,
+      "loss": 1.8141,
+      "step": 104
+    },
+    {
+      "epoch": 0.10785824345146379,
+      "grad_norm": 2.2768120765686035,
+      "learning_rate": 1.5239726027397262e-06,
+      "loss": 1.8826,
+      "step": 105
+    },
+    {
+      "epoch": 0.10888546481766821,
+      "grad_norm": 2.316584348678589,
+      "learning_rate": 1.541095890410959e-06,
+      "loss": 1.8604,
+      "step": 106
+    },
+    {
+      "epoch": 0.10991268618387262,
+      "grad_norm": 2.14679217338562,
+      "learning_rate": 1.5582191780821921e-06,
+      "loss": 1.865,
+      "step": 107
+    },
+    {
+      "epoch": 0.11093990755007704,
+      "grad_norm": 2.3288021087646484,
+      "learning_rate": 1.5753424657534248e-06,
+      "loss": 1.8512,
+      "step": 108
+    },
+    {
+      "epoch": 0.11196712891628145,
+      "grad_norm": 2.300482988357544,
+      "learning_rate": 1.5924657534246576e-06,
+      "loss": 1.7455,
+      "step": 109
+    },
+    {
+      "epoch": 0.11299435028248588,
+      "grad_norm": 2.4226229190826416,
+      "learning_rate": 1.6095890410958907e-06,
+      "loss": 1.8058,
+      "step": 110
+    },
+    {
+      "epoch": 0.11299435028248588,
+      "eval_loss": 2.025261640548706,
+      "eval_runtime": 37.6779,
+      "eval_samples_per_second": 26.541,
+      "eval_steps_per_second": 1.115,
+      "step": 110
+    },
+    {
+      "epoch": 0.1140215716486903,
+      "grad_norm": 2.0930111408233643,
+      "learning_rate": 1.6267123287671235e-06,
+      "loss": 1.7375,
+      "step": 111
+    },
+    {
+      "epoch": 0.1150487930148947,
+      "grad_norm": 2.4973645210266113,
+      "learning_rate": 1.6438356164383561e-06,
+      "loss": 1.8268,
+      "step": 112
+    },
+    {
+      "epoch": 0.11607601438109913,
+      "grad_norm": 2.3295114040374756,
+      "learning_rate": 1.6609589041095892e-06,
+      "loss": 1.7882,
+      "step": 113
+    },
+    {
+      "epoch": 0.11710323574730354,
+      "grad_norm": 2.3693318367004395,
+      "learning_rate": 1.678082191780822e-06,
+      "loss": 1.8192,
+      "step": 114
+    },
+    {
+      "epoch": 0.11813045711350796,
+      "grad_norm": 2.553637742996216,
+      "learning_rate": 1.695205479452055e-06,
+      "loss": 1.7491,
+      "step": 115
+    },
+    {
+      "epoch": 0.11915767847971238,
+      "grad_norm": 2.303945541381836,
+      "learning_rate": 1.7123287671232877e-06,
+      "loss": 1.8737,
+      "step": 116
+    },
+    {
+      "epoch": 0.12018489984591679,
+      "grad_norm": 2.250356435775757,
+      "learning_rate": 1.7294520547945206e-06,
+      "loss": 1.8039,
+      "step": 117
+    },
+    {
+      "epoch": 0.12121212121212122,
+      "grad_norm": 2.3854403495788574,
+      "learning_rate": 1.7465753424657536e-06,
+      "loss": 1.8357,
+      "step": 118
+    },
+    {
+      "epoch": 0.12223934257832562,
+      "grad_norm": 2.4049572944641113,
+      "learning_rate": 1.7636986301369865e-06,
+      "loss": 1.8209,
+      "step": 119
+    },
+    {
+      "epoch": 0.12326656394453005,
+      "grad_norm": 2.313551425933838,
+      "learning_rate": 1.7808219178082193e-06,
+      "loss": 1.9026,
+      "step": 120
+    },
+    {
+      "epoch": 0.12326656394453005,
+      "eval_loss": 2.012171745300293,
+      "eval_runtime": 37.7908,
+      "eval_samples_per_second": 26.461,
+      "eval_steps_per_second": 1.111,
+      "step": 120
+    },
+    {
+      "epoch": 0.12429378531073447,
+      "grad_norm": 2.518320322036743,
+      "learning_rate": 1.7979452054794521e-06,
+      "loss": 1.8241,
+      "step": 121
+    },
+    {
+      "epoch": 0.12532100667693888,
+      "grad_norm": 2.6504592895507812,
+      "learning_rate": 1.8150684931506852e-06,
+      "loss": 1.8091,
+      "step": 122
+    },
+    {
+      "epoch": 0.1263482280431433,
+      "grad_norm": 2.556610107421875,
+      "learning_rate": 1.832191780821918e-06,
+      "loss": 1.7547,
+      "step": 123
+    },
+    {
+      "epoch": 0.12737544940934772,
+      "grad_norm": 2.247633695602417,
+      "learning_rate": 1.8493150684931507e-06,
+      "loss": 1.7923,
+      "step": 124
+    },
+    {
+      "epoch": 0.12840267077555212,
+      "grad_norm": 2.195596933364868,
+      "learning_rate": 1.8664383561643837e-06,
+      "loss": 1.747,
+      "step": 125
+    },
+    {
+      "epoch": 0.12942989214175654,
+      "grad_norm": 2.121396780014038,
+      "learning_rate": 1.8835616438356166e-06,
+      "loss": 1.8088,
+      "step": 126
+    },
+    {
+      "epoch": 0.13045711350796096,
+      "grad_norm": 2.4572501182556152,
+      "learning_rate": 1.9006849315068496e-06,
+      "loss": 1.7285,
+      "step": 127
+    },
+    {
+      "epoch": 0.1314843348741654,
+      "grad_norm": 2.2590177059173584,
+      "learning_rate": 1.9178082191780823e-06,
+      "loss": 1.7929,
+      "step": 128
+    },
+    {
+      "epoch": 0.1325115562403698,
+      "grad_norm": 2.5099234580993652,
+      "learning_rate": 1.9349315068493153e-06,
+      "loss": 1.8365,
+      "step": 129
+    },
+    {
+      "epoch": 0.1335387776065742,
+      "grad_norm": 2.3064465522766113,
+      "learning_rate": 1.952054794520548e-06,
+      "loss": 1.8351,
+      "step": 130
+    },
+    {
+      "epoch": 0.1335387776065742,
+      "eval_loss": 2.0043866634368896,
+      "eval_runtime": 37.8661,
+      "eval_samples_per_second": 26.409,
+      "eval_steps_per_second": 1.109,
+      "step": 130
+    },
+    {
+      "epoch": 0.13456599897277863,
+      "grad_norm": 2.5977718830108643,
+      "learning_rate": 1.969178082191781e-06,
+      "loss": 1.7603,
+      "step": 131
+    },
+    {
+      "epoch": 0.13559322033898305,
+      "grad_norm": 2.23221755027771,
+      "learning_rate": 1.9863013698630136e-06,
+      "loss": 1.8237,
+      "step": 132
+    },
+    {
+      "epoch": 0.13662044170518747,
+      "grad_norm": 2.247044324874878,
+      "learning_rate": 2.0034246575342467e-06,
+      "loss": 1.8006,
+      "step": 133
+    },
+    {
+      "epoch": 0.1376476630713919,
+      "grad_norm": 2.1796107292175293,
+      "learning_rate": 2.0205479452054797e-06,
+      "loss": 1.7354,
+      "step": 134
+    },
+    {
+      "epoch": 0.1386748844375963,
+      "grad_norm": 2.3269693851470947,
+      "learning_rate": 2.037671232876713e-06,
+      "loss": 1.783,
+      "step": 135
+    },
+    {
+      "epoch": 0.1397021058038007,
+      "grad_norm": 2.2903053760528564,
+      "learning_rate": 2.0547945205479454e-06,
+      "loss": 1.8072,
+      "step": 136
+    },
+    {
+      "epoch": 0.14072932717000514,
+      "grad_norm": 2.634150981903076,
+      "learning_rate": 2.071917808219178e-06,
+      "loss": 1.7251,
+      "step": 137
+    },
+    {
+      "epoch": 0.14175654853620956,
+      "grad_norm": 2.42579984664917,
+      "learning_rate": 2.089041095890411e-06,
+      "loss": 1.7997,
+      "step": 138
+    },
+    {
+      "epoch": 0.14278376990241398,
+      "grad_norm": 2.615828514099121,
+      "learning_rate": 2.106164383561644e-06,
+      "loss": 1.766,
+      "step": 139
+    },
+    {
+      "epoch": 0.14381099126861838,
+      "grad_norm": 2.312714099884033,
+      "learning_rate": 2.123287671232877e-06,
+      "loss": 1.7538,
+      "step": 140
+    },
+    {
+      "epoch": 0.14381099126861838,
+      "eval_loss": 1.9918214082717896,
+      "eval_runtime": 37.8698,
+      "eval_samples_per_second": 26.406,
+      "eval_steps_per_second": 1.109,
+      "step": 140
+    },
+    {
+      "epoch": 0.1448382126348228,
+      "grad_norm": 2.7540030479431152,
+      "learning_rate": 2.14041095890411e-06,
+      "loss": 1.7856,
+      "step": 141
+    },
+    {
+      "epoch": 0.14586543400102722,
+      "grad_norm": 2.289191246032715,
+      "learning_rate": 2.1575342465753425e-06,
+      "loss": 1.7288,
+      "step": 142
+    },
+    {
+      "epoch": 0.14689265536723164,
+      "grad_norm": 2.788001775741577,
+      "learning_rate": 2.1746575342465755e-06,
+      "loss": 1.8703,
+      "step": 143
+    },
+    {
+      "epoch": 0.14791987673343607,
+      "grad_norm": 2.4128854274749756,
+      "learning_rate": 2.191780821917808e-06,
+      "loss": 1.7461,
+      "step": 144
+    },
+    {
+      "epoch": 0.14894709809964046,
+      "grad_norm": 2.4108099937438965,
+      "learning_rate": 2.2089041095890412e-06,
+      "loss": 1.73,
+      "step": 145
+    },
+    {
+      "epoch": 0.14997431946584489,
+      "grad_norm": 2.4315948486328125,
+      "learning_rate": 2.2260273972602743e-06,
+      "loss": 1.7303,
+      "step": 146
+    },
+    {
+      "epoch": 0.1510015408320493,
+      "grad_norm": 2.494056224822998,
+      "learning_rate": 2.243150684931507e-06,
+      "loss": 1.7616,
+      "step": 147
+    },
+    {
+      "epoch": 0.15202876219825373,
+      "grad_norm": 2.4557394981384277,
+      "learning_rate": 2.26027397260274e-06,
+      "loss": 1.7655,
+      "step": 148
+    },
+    {
+      "epoch": 0.15305598356445815,
+      "grad_norm": 2.4114091396331787,
+      "learning_rate": 2.2773972602739726e-06,
+      "loss": 1.8538,
+      "step": 149
+    },
+    {
+      "epoch": 0.15408320493066255,
+      "grad_norm": 2.609219789505005,
+      "learning_rate": 2.2945205479452057e-06,
+      "loss": 1.7767,
+      "step": 150
+    },
+    {
+      "epoch": 0.15408320493066255,
+      "eval_loss": 1.981367588043213,
+      "eval_runtime": 37.9562,
+      "eval_samples_per_second": 26.346,
+      "eval_steps_per_second": 1.107,
+      "step": 150
+    },
+    {
+      "epoch": 0.15511042629686697,
+      "grad_norm": 2.213622808456421,
+      "learning_rate": 2.3116438356164387e-06,
+      "loss": 1.687,
+      "step": 151
+    },
+    {
+      "epoch": 0.1561376476630714,
+      "grad_norm": 2.6517975330352783,
+      "learning_rate": 2.3287671232876713e-06,
+      "loss": 1.7311,
+      "step": 152
+    },
+    {
+      "epoch": 0.15716486902927582,
+      "grad_norm": 2.5174458026885986,
+      "learning_rate": 2.3458904109589044e-06,
+      "loss": 1.7835,
+      "step": 153
+    },
+    {
+      "epoch": 0.15819209039548024,
+      "grad_norm": 2.485480308532715,
+      "learning_rate": 2.363013698630137e-06,
+      "loss": 1.7749,
+      "step": 154
+    },
+    {
+      "epoch": 0.15921931176168463,
+      "grad_norm": 2.5386061668395996,
+      "learning_rate": 2.38013698630137e-06,
+      "loss": 1.7223,
+      "step": 155
+    },
+    {
+      "epoch": 0.16024653312788906,
+      "grad_norm": 2.2895941734313965,
+      "learning_rate": 2.3972602739726027e-06,
+      "loss": 1.6845,
+      "step": 156
+    },
+    {
+      "epoch": 0.16127375449409348,
+      "grad_norm": 2.5453789234161377,
+      "learning_rate": 2.4143835616438358e-06,
+      "loss": 1.7627,
+      "step": 157
+    },
+    {
+      "epoch": 0.1623009758602979,
+      "grad_norm": 2.5322303771972656,
+      "learning_rate": 2.431506849315069e-06,
+      "loss": 1.764,
+      "step": 158
+    },
+    {
+      "epoch": 0.1633281972265023,
+      "grad_norm": 2.2669899463653564,
+      "learning_rate": 2.4486301369863015e-06,
+      "loss": 1.7487,
+      "step": 159
+    },
+    {
+      "epoch": 0.16435541859270672,
+      "grad_norm": 2.4577853679656982,
+      "learning_rate": 2.4657534246575345e-06,
+      "loss": 1.7635,
+      "step": 160
+    },
+    {
+      "epoch": 0.16435541859270672,
+      "eval_loss": 1.9793355464935303,
+      "eval_runtime": 37.8576,
+      "eval_samples_per_second": 26.415,
+      "eval_steps_per_second": 1.109,
+      "step": 160
+    },
+    {
+      "epoch": 0.16538263995891114,
+      "grad_norm": 2.354893922805786,
+      "learning_rate": 2.482876712328767e-06,
+      "loss": 1.7538,
+      "step": 161
+    },
+    {
+      "epoch": 0.16640986132511557,
+      "grad_norm": 2.3255856037139893,
+      "learning_rate": 2.5e-06,
+      "loss": 1.7277,
+      "step": 162
+    },
+    {
+      "epoch": 0.16743708269132,
+      "grad_norm": 2.3501994609832764,
+      "learning_rate": 2.5171232876712333e-06,
+      "loss": 1.6985,
+      "step": 163
+    },
+    {
+      "epoch": 0.16846430405752438,
+      "grad_norm": 2.386627674102783,
+      "learning_rate": 2.534246575342466e-06,
+      "loss": 1.7552,
+      "step": 164
+    },
+    {
+      "epoch": 0.1694915254237288,
+      "grad_norm": 2.2238457202911377,
+      "learning_rate": 2.551369863013699e-06,
+      "loss": 1.7805,
+      "step": 165
+    },
+    {
+      "epoch": 0.17051874678993323,
+      "grad_norm": 2.2653403282165527,
+      "learning_rate": 2.568493150684932e-06,
+      "loss": 1.7656,
+      "step": 166
+    },
+    {
+      "epoch": 0.17154596815613765,
+      "grad_norm": 2.233132839202881,
+      "learning_rate": 2.585616438356164e-06,
+      "loss": 1.6997,
+      "step": 167
+    },
+    {
+      "epoch": 0.17257318952234207,
+      "grad_norm": 2.1696043014526367,
+      "learning_rate": 2.6027397260273973e-06,
+      "loss": 1.7487,
+      "step": 168
+    },
+    {
+      "epoch": 0.17360041088854647,
+      "grad_norm": 2.4479899406433105,
+      "learning_rate": 2.6198630136986303e-06,
+      "loss": 1.7957,
+      "step": 169
+    },
+    {
+      "epoch": 0.1746276322547509,
+      "grad_norm": 2.3054702281951904,
+      "learning_rate": 2.6369863013698634e-06,
+      "loss": 1.766,
+      "step": 170
+    },
+    {
+      "epoch": 0.1746276322547509,
+      "eval_loss": 1.9543790817260742,
+      "eval_runtime": 37.8837,
+      "eval_samples_per_second": 26.397,
+      "eval_steps_per_second": 1.109,
+      "step": 170
+    },
+    {
+      "epoch": 0.17565485362095531,
+      "grad_norm": 2.3591110706329346,
+      "learning_rate": 2.654109589041096e-06,
+      "loss": 1.7149,
+      "step": 171
+    },
+    {
+      "epoch": 0.17668207498715974,
+      "grad_norm": 2.450526237487793,
+      "learning_rate": 2.671232876712329e-06,
+      "loss": 1.7074,
+      "step": 172
+    },
+    {
+      "epoch": 0.17770929635336416,
+      "grad_norm": 2.1511754989624023,
+      "learning_rate": 2.688356164383562e-06,
+      "loss": 1.7036,
+      "step": 173
+    },
+    {
+      "epoch": 0.17873651771956856,
+      "grad_norm": 2.5944275856018066,
+      "learning_rate": 2.705479452054795e-06,
+      "loss": 1.6566,
+      "step": 174
+    },
+    {
+      "epoch": 0.17976373908577298,
+      "grad_norm": 2.4301369190216064,
+      "learning_rate": 2.7226027397260274e-06,
+      "loss": 1.7977,
+      "step": 175
+    },
+    {
+      "epoch": 0.1807909604519774,
+      "grad_norm": 2.287665367126465,
+      "learning_rate": 2.7397260273972604e-06,
+      "loss": 1.7526,
+      "step": 176
+    },
+    {
+      "epoch": 0.18181818181818182,
+      "grad_norm": 2.380032777786255,
+      "learning_rate": 2.7568493150684935e-06,
+      "loss": 1.6753,
+      "step": 177
+    },
+    {
+      "epoch": 0.18284540318438625,
+      "grad_norm": 2.4040510654449463,
+      "learning_rate": 2.773972602739726e-06,
+      "loss": 1.758,
+      "step": 178
+    },
+    {
+      "epoch": 0.18387262455059064,
+      "grad_norm": 2.5074045658111572,
+      "learning_rate": 2.791095890410959e-06,
+      "loss": 1.6498,
+      "step": 179
+    },
+    {
+      "epoch": 0.18489984591679506,
+      "grad_norm": 2.3926949501037598,
+      "learning_rate": 2.8082191780821922e-06,
+      "loss": 1.8006,
+      "step": 180
+    },
+    {
+      "epoch": 0.18489984591679506,
+      "eval_loss": 1.9446762800216675,
+      "eval_runtime": 37.8572,
+      "eval_samples_per_second": 26.415,
+      "eval_steps_per_second": 1.109,
+      "step": 180
+    },
+    {
+      "epoch": 0.1859270672829995,
+      "grad_norm": 2.4991633892059326,
+      "learning_rate": 2.8253424657534253e-06,
+      "loss": 1.7649,
+      "step": 181
+    },
+    {
+      "epoch": 0.1869542886492039,
+      "grad_norm": 2.2558324337005615,
+      "learning_rate": 2.842465753424658e-06,
+      "loss": 1.7743,
+      "step": 182
+    },
+    {
+      "epoch": 0.18798151001540833,
+      "grad_norm": 2.219717025756836,
+      "learning_rate": 2.8595890410958905e-06,
+      "loss": 1.7807,
+      "step": 183
+    },
+    {
+      "epoch": 0.18900873138161273,
+      "grad_norm": 2.406562566757202,
+      "learning_rate": 2.876712328767123e-06,
+      "loss": 1.6289,
+      "step": 184
+    },
+    {
+      "epoch": 0.19003595274781715,
+      "grad_norm": 2.356579303741455,
+      "learning_rate": 2.8938356164383562e-06,
+      "loss": 1.6747,
+      "step": 185
+    },
+    {
+      "epoch": 0.19106317411402157,
+      "grad_norm": 2.4844322204589844,
+      "learning_rate": 2.9109589041095893e-06,
+      "loss": 1.6731,
+      "step": 186
+    },
+    {
+      "epoch": 0.192090395480226,
+      "grad_norm": 2.8179149627685547,
+      "learning_rate": 2.9280821917808223e-06,
+      "loss": 1.6786,
+      "step": 187
+    },
+    {
+      "epoch": 0.19311761684643042,
+      "grad_norm": 2.395416259765625,
+      "learning_rate": 2.945205479452055e-06,
+      "loss": 1.7313,
+      "step": 188
+    },
+    {
+      "epoch": 0.1941448382126348,
+      "grad_norm": 2.848740816116333,
+      "learning_rate": 2.962328767123288e-06,
+      "loss": 1.7258,
+      "step": 189
+    },
+    {
+      "epoch": 0.19517205957883924,
+      "grad_norm": 2.427395820617676,
+      "learning_rate": 2.979452054794521e-06,
+      "loss": 1.6633,
+      "step": 190
+    },
+    {
+      "epoch": 0.19517205957883924,
+      "eval_loss": 1.9443999528884888,
+      "eval_runtime": 37.8579,
+      "eval_samples_per_second": 26.415,
+      "eval_steps_per_second": 1.109,
+      "step": 190
+    },
+    {
+      "epoch": 0.19619928094504366,
+      "grad_norm": 2.5347256660461426,
+      "learning_rate": 2.9965753424657533e-06,
+      "loss": 1.7342,
+      "step": 191
+    },
+    {
+      "epoch": 0.19722650231124808,
+      "grad_norm": 2.8696866035461426,
+      "learning_rate": 3.0136986301369864e-06,
+      "loss": 1.7037,
+      "step": 192
+    },
+    {
+      "epoch": 0.1982537236774525,
+      "grad_norm": 2.427468776702881,
+      "learning_rate": 3.0308219178082194e-06,
+      "loss": 1.672,
+      "step": 193
+    },
+    {
+      "epoch": 0.1992809450436569,
+      "grad_norm": 2.4253737926483154,
+      "learning_rate": 3.0479452054794525e-06,
+      "loss": 1.7252,
+      "step": 194
+    },
+    {
+      "epoch": 0.20030816640986132,
+      "grad_norm": 2.801668643951416,
+      "learning_rate": 3.065068493150685e-06,
+      "loss": 1.6694,
+      "step": 195
+    },
+    {
+      "epoch": 0.20133538777606574,
+      "grad_norm": 2.442263603210449,
+      "learning_rate": 3.082191780821918e-06,
+      "loss": 1.6866,
+      "step": 196
+    },
+    {
+      "epoch": 0.20236260914227017,
+      "grad_norm": 2.777463674545288,
+      "learning_rate": 3.099315068493151e-06,
+      "loss": 1.686,
+      "step": 197
+    },
+    {
+      "epoch": 0.2033898305084746,
+      "grad_norm": 2.491227388381958,
+      "learning_rate": 3.1164383561643843e-06,
+      "loss": 1.6414,
+      "step": 198
+    },
+    {
+      "epoch": 0.20441705187467898,
+      "grad_norm": 2.33585262298584,
+      "learning_rate": 3.1335616438356165e-06,
+      "loss": 1.7808,
+      "step": 199
+    },
+    {
+      "epoch": 0.2054442732408834,
+      "grad_norm": 2.518160581588745,
+      "learning_rate": 3.1506849315068495e-06,
+      "loss": 1.6592,
+      "step": 200
+    },
+    {
+      "epoch": 0.2054442732408834,
+      "eval_loss": 1.9371097087860107,
+      "eval_runtime": 37.8602,
+      "eval_samples_per_second": 26.413,
+      "eval_steps_per_second": 1.109,
+      "step": 200
+    },
+    {
+      "epoch": 0.20647149460708783,
+      "grad_norm": 2.460204839706421,
+      "learning_rate": 3.167808219178082e-06,
+      "loss": 1.6668,
+      "step": 201
+    },
+    {
+      "epoch": 0.20749871597329225,
+      "grad_norm": 2.377824068069458,
+      "learning_rate": 3.184931506849315e-06,
+      "loss": 1.6462,
+      "step": 202
+    },
+    {
+      "epoch": 0.20852593733949668,
+      "grad_norm": 2.426177740097046,
+      "learning_rate": 3.2020547945205483e-06,
+      "loss": 1.6272,
+      "step": 203
+    },
+    {
+      "epoch": 0.20955315870570107,
+      "grad_norm": 2.4532034397125244,
+      "learning_rate": 3.2191780821917813e-06,
+      "loss": 1.6169,
+      "step": 204
+    },
+    {
+      "epoch": 0.2105803800719055,
+      "grad_norm": 2.517059564590454,
+      "learning_rate": 3.236301369863014e-06,
+      "loss": 1.5991,
+      "step": 205
+    },
+    {
+      "epoch": 0.21160760143810992,
+      "grad_norm": 2.22658109664917,
+      "learning_rate": 3.253424657534247e-06,
+      "loss": 1.6754,
+      "step": 206
+    },
+    {
+      "epoch": 0.21263482280431434,
+      "grad_norm": 2.344799518585205,
+      "learning_rate": 3.2705479452054796e-06,
+      "loss": 1.6648,
+      "step": 207
+    },
+    {
+      "epoch": 0.21366204417051873,
+      "grad_norm": 2.331875801086426,
+      "learning_rate": 3.2876712328767123e-06,
+      "loss": 1.6112,
+      "step": 208
+    },
+    {
+      "epoch": 0.21468926553672316,
+      "grad_norm": 2.262343645095825,
+      "learning_rate": 3.3047945205479453e-06,
+      "loss": 1.6479,
+      "step": 209
+    },
+    {
+      "epoch": 0.21571648690292758,
+      "grad_norm": 2.282278537750244,
+      "learning_rate": 3.3219178082191784e-06,
+      "loss": 1.6753,
+      "step": 210
+    },
+    {
+      "epoch": 0.21571648690292758,
+      "eval_loss": 1.9338418245315552,
+      "eval_runtime": 37.6083,
+      "eval_samples_per_second": 26.59,
+      "eval_steps_per_second": 1.117,
+      "step": 210
+    },
+    {
+      "epoch": 0.216743708269132,
+      "grad_norm": 2.578476905822754,
+      "learning_rate": 3.3390410958904114e-06,
+      "loss": 1.7043,
+      "step": 211
+    },
+    {
+      "epoch": 0.21777092963533642,
+      "grad_norm": 2.3118813037872314,
+      "learning_rate": 3.356164383561644e-06,
+      "loss": 1.7541,
+      "step": 212
+    },
+    {
+      "epoch": 0.21879815100154082,
+      "grad_norm": 2.2418692111968994,
+      "learning_rate": 3.373287671232877e-06,
+      "loss": 1.6886,
+      "step": 213
+    },
+    {
+      "epoch": 0.21982537236774524,
+      "grad_norm": 2.3583438396453857,
+      "learning_rate": 3.39041095890411e-06,
+      "loss": 1.7241,
+      "step": 214
+    },
+    {
+      "epoch": 0.22085259373394966,
+      "grad_norm": 2.345090866088867,
+      "learning_rate": 3.4075342465753424e-06,
+      "loss": 1.6194,
+      "step": 215
+    },
+    {
+      "epoch": 0.2218798151001541,
+      "grad_norm": 2.3996200561523438,
+      "learning_rate": 3.4246575342465754e-06,
+      "loss": 1.6889,
+      "step": 216
+    },
+    {
+      "epoch": 0.2229070364663585,
+      "grad_norm": 2.3552374839782715,
+      "learning_rate": 3.4417808219178085e-06,
+      "loss": 1.6713,
+      "step": 217
+    },
+    {
+      "epoch": 0.2239342578325629,
+      "grad_norm": 2.2653744220733643,
+      "learning_rate": 3.458904109589041e-06,
+      "loss": 1.673,
+      "step": 218
+    },
+    {
+      "epoch": 0.22496147919876733,
+      "grad_norm": 2.263288974761963,
+      "learning_rate": 3.476027397260274e-06,
+      "loss": 1.673,
+      "step": 219
+    },
+    {
+      "epoch": 0.22598870056497175,
+      "grad_norm": 2.3173880577087402,
+      "learning_rate": 3.4931506849315072e-06,
+      "loss": 1.6803,
+      "step": 220
+    },
+    {
+      "epoch": 0.22598870056497175,
+      "eval_loss": 1.9166476726531982,
+      "eval_runtime": 37.7659,
+      "eval_samples_per_second": 26.479,
+      "eval_steps_per_second": 1.112,
+      "step": 220
+    },
+    {
+      "epoch": 0.22701592193117617,
+      "grad_norm": 2.4005839824676514,
+      "learning_rate": 3.5102739726027403e-06,
+      "loss": 1.6809,
+      "step": 221
+    },
+    {
+      "epoch": 0.2280431432973806,
+      "grad_norm": 2.5032525062561035,
+      "learning_rate": 3.527397260273973e-06,
+      "loss": 1.6411,
+      "step": 222
+    },
+    {
+      "epoch": 0.229070364663585,
+      "grad_norm": 2.533583879470825,
+      "learning_rate": 3.5445205479452056e-06,
+      "loss": 1.6608,
+      "step": 223
+    },
+    {
+      "epoch": 0.2300975860297894,
+      "grad_norm": 2.3575572967529297,
+      "learning_rate": 3.5616438356164386e-06,
+      "loss": 1.7236,
+      "step": 224
+    },
+    {
+      "epoch": 0.23112480739599384,
+      "grad_norm": 2.3408942222595215,
+      "learning_rate": 3.5787671232876712e-06,
+      "loss": 1.7101,
+      "step": 225
+    },
+    {
+      "epoch": 0.23215202876219826,
+      "grad_norm": 2.3819596767425537,
+      "learning_rate": 3.5958904109589043e-06,
+      "loss": 1.6414,
+      "step": 226
+    },
+    {
+      "epoch": 0.23317925012840268,
+      "grad_norm": 2.2874338626861572,
+      "learning_rate": 3.6130136986301373e-06,
+      "loss": 1.6767,
+      "step": 227
+    },
+    {
+      "epoch": 0.23420647149460708,
+      "grad_norm": 2.3523948192596436,
+      "learning_rate": 3.6301369863013704e-06,
+      "loss": 1.6932,
+      "step": 228
+    },
+    {
+      "epoch": 0.2352336928608115,
+      "grad_norm": 2.476128578186035,
+      "learning_rate": 3.647260273972603e-06,
+      "loss": 1.7125,
+      "step": 229
+    },
+    {
+      "epoch": 0.23626091422701592,
+      "grad_norm": 2.3841593265533447,
+      "learning_rate": 3.664383561643836e-06,
+      "loss": 1.6469,
+      "step": 230
+    },
+    {
+      "epoch": 0.23626091422701592,
+      "eval_loss": 1.925522804260254,
+      "eval_runtime": 37.8581,
+      "eval_samples_per_second": 26.414,
+      "eval_steps_per_second": 1.109,
+      "step": 230
+    },
+    {
+      "epoch": 0.23728813559322035,
+      "grad_norm": 2.4401519298553467,
+      "learning_rate": 3.6815068493150687e-06,
+      "loss": 1.6726,
+      "step": 231
+    },
+    {
+      "epoch": 0.23831535695942477,
+      "grad_norm": 2.329674005508423,
+      "learning_rate": 3.6986301369863014e-06,
+      "loss": 1.6349,
+      "step": 232
+    },
+    {
+      "epoch": 0.23934257832562916,
+      "grad_norm": 2.654290199279785,
+      "learning_rate": 3.7157534246575344e-06,
+      "loss": 1.7309,
+      "step": 233
+    },
+    {
+      "epoch": 0.24036979969183359,
+      "grad_norm": 2.1599373817443848,
+      "learning_rate": 3.7328767123287675e-06,
+      "loss": 1.6437,
+      "step": 234
+    },
+    {
+      "epoch": 0.241397021058038,
+      "grad_norm": 2.6170263290405273,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 1.6862,
+      "step": 235
+    },
+    {
+      "epoch": 0.24242424242424243,
+      "grad_norm": 2.2836415767669678,
+      "learning_rate": 3.767123287671233e-06,
+      "loss": 1.7073,
+      "step": 236
+    },
+    {
+      "epoch": 0.24345146379044685,
+      "grad_norm": 2.3001456260681152,
+      "learning_rate": 3.784246575342466e-06,
+      "loss": 1.6455,
+      "step": 237
+    },
+    {
+      "epoch": 0.24447868515665125,
+      "grad_norm": 2.4549131393432617,
+      "learning_rate": 3.8013698630136993e-06,
+      "loss": 1.6566,
+      "step": 238
+    },
+    {
+      "epoch": 0.24550590652285567,
+      "grad_norm": 2.4387831687927246,
+      "learning_rate": 3.818493150684932e-06,
+      "loss": 1.627,
+      "step": 239
+    },
+    {
+      "epoch": 0.2465331278890601,
+      "grad_norm": 2.136474847793579,
+      "learning_rate": 3.8356164383561645e-06,
+      "loss": 1.7286,
+      "step": 240
+    },
+    {
+      "epoch": 0.2465331278890601,
+      "eval_loss": 1.9188542366027832,
+      "eval_runtime": 37.9079,
+      "eval_samples_per_second": 26.38,
+      "eval_steps_per_second": 1.108,
+      "step": 240
+    },
+    {
+      "epoch": 0.24756034925526452,
+      "grad_norm": 2.430297374725342,
+      "learning_rate": 3.852739726027397e-06,
+      "loss": 1.7014,
+      "step": 241
+    },
+    {
+      "epoch": 0.24858757062146894,
+      "grad_norm": 2.302480697631836,
+      "learning_rate": 3.869863013698631e-06,
+      "loss": 1.6412,
+      "step": 242
+    },
+    {
+      "epoch": 0.24961479198767333,
+      "grad_norm": 2.2905426025390625,
+      "learning_rate": 3.886986301369863e-06,
+      "loss": 1.6445,
+      "step": 243
+    },
+    {
+      "epoch": 0.25064201335387776,
+      "grad_norm": 2.55385684967041,
+      "learning_rate": 3.904109589041096e-06,
+      "loss": 1.6265,
+      "step": 244
+    },
+    {
+      "epoch": 0.2516692347200822,
+      "grad_norm": 2.3053576946258545,
+      "learning_rate": 3.921232876712329e-06,
+      "loss": 1.6338,
+      "step": 245
+    },
+    {
+      "epoch": 0.2526964560862866,
+      "grad_norm": 2.4894707202911377,
+      "learning_rate": 3.938356164383562e-06,
+      "loss": 1.5966,
+      "step": 246
+    },
+    {
+      "epoch": 0.253723677452491,
+      "grad_norm": 2.6198930740356445,
+      "learning_rate": 3.9554794520547955e-06,
+      "loss": 1.6935,
+      "step": 247
+    },
+    {
+      "epoch": 0.25475089881869545,
+      "grad_norm": 2.3548998832702637,
+      "learning_rate": 3.972602739726027e-06,
+      "loss": 1.6227,
+      "step": 248
+    },
+    {
+      "epoch": 0.25577812018489987,
+      "grad_norm": 2.535460948944092,
+      "learning_rate": 3.989726027397261e-06,
+      "loss": 1.6699,
+      "step": 249
+    },
+    {
+      "epoch": 0.25680534155110424,
+      "grad_norm": 2.3680567741394043,
+      "learning_rate": 4.006849315068493e-06,
+      "loss": 1.6666,
+      "step": 250
+    },
+    {
+      "epoch": 0.25680534155110424,
+      "eval_loss": 1.9149967432022095,
+      "eval_runtime": 37.9423,
+      "eval_samples_per_second": 26.356,
+      "eval_steps_per_second": 1.107,
+      "step": 250
+    },
+    {
+      "epoch": 0.25783256291730866,
+      "grad_norm": 2.3914358615875244,
+      "learning_rate": 4.023972602739726e-06,
+      "loss": 1.6633,
+      "step": 251
+    },
+    {
+      "epoch": 0.2588597842835131,
+      "grad_norm": 2.4293038845062256,
+      "learning_rate": 4.0410958904109595e-06,
+      "loss": 1.6573,
+      "step": 252
+    },
+    {
+      "epoch": 0.2598870056497175,
+      "grad_norm": 2.3409900665283203,
+      "learning_rate": 4.058219178082192e-06,
+      "loss": 1.6815,
+      "step": 253
+    },
+    {
+      "epoch": 0.26091422701592193,
+      "grad_norm": 2.5807456970214844,
+      "learning_rate": 4.075342465753426e-06,
+      "loss": 1.6412,
+      "step": 254
+    },
+    {
+      "epoch": 0.26194144838212635,
+      "grad_norm": 2.317861795425415,
+      "learning_rate": 4.092465753424658e-06,
+      "loss": 1.6311,
+      "step": 255
+    },
+    {
+      "epoch": 0.2629686697483308,
+      "grad_norm": 2.231682777404785,
+      "learning_rate": 4.109589041095891e-06,
+      "loss": 1.7021,
+      "step": 256
+    },
+    {
+      "epoch": 0.2639958911145352,
+      "grad_norm": 2.7119040489196777,
+      "learning_rate": 4.1267123287671235e-06,
+      "loss": 1.6446,
+      "step": 257
+    },
+    {
+      "epoch": 0.2650231124807396,
+      "grad_norm": 2.2619071006774902,
+      "learning_rate": 4.143835616438356e-06,
+      "loss": 1.7101,
+      "step": 258
+    },
+    {
+      "epoch": 0.26605033384694404,
+      "grad_norm": 2.461945056915283,
+      "learning_rate": 4.16095890410959e-06,
+      "loss": 1.6901,
+      "step": 259
+    },
+    {
+      "epoch": 0.2670775552131484,
+      "grad_norm": 2.516348123550415,
+      "learning_rate": 4.178082191780822e-06,
+      "loss": 1.644,
+      "step": 260
+    },
+    {
+      "epoch": 0.2670775552131484,
+      "eval_loss": 1.904897928237915,
+      "eval_runtime": 37.93,
+      "eval_samples_per_second": 26.364,
+      "eval_steps_per_second": 1.107,
+      "step": 260
+    },
+    {
+      "epoch": 0.26810477657935283,
+      "grad_norm": 2.2311813831329346,
+      "learning_rate": 4.195205479452055e-06,
+      "loss": 1.6301,
+      "step": 261
+    },
+    {
+      "epoch": 0.26913199794555726,
+      "grad_norm": 2.605109214782715,
+      "learning_rate": 4.212328767123288e-06,
+      "loss": 1.6393,
+      "step": 262
+    },
+    {
+      "epoch": 0.2701592193117617,
+      "grad_norm": 2.6103134155273438,
+      "learning_rate": 4.229452054794521e-06,
+      "loss": 1.6468,
+      "step": 263
+    },
+    {
+      "epoch": 0.2711864406779661,
+      "grad_norm": 2.5538523197174072,
+      "learning_rate": 4.246575342465754e-06,
+      "loss": 1.6294,
+      "step": 264
+    },
+    {
+      "epoch": 0.2722136620441705,
+      "grad_norm": 2.5841965675354004,
+      "learning_rate": 4.263698630136986e-06,
+      "loss": 1.5837,
+      "step": 265
+    },
+    {
+      "epoch": 0.27324088341037495,
+      "grad_norm": 2.4604523181915283,
+      "learning_rate": 4.28082191780822e-06,
+      "loss": 1.7473,
+      "step": 266
+    },
+    {
+      "epoch": 0.27426810477657937,
+      "grad_norm": 2.6701765060424805,
+      "learning_rate": 4.297945205479452e-06,
+      "loss": 1.6539,
+      "step": 267
+    },
+    {
+      "epoch": 0.2752953261427838,
+      "grad_norm": 2.360069990158081,
+      "learning_rate": 4.315068493150685e-06,
+      "loss": 1.6385,
+      "step": 268
+    },
+    {
+      "epoch": 0.2763225475089882,
+      "grad_norm": 2.4922797679901123,
+      "learning_rate": 4.3321917808219185e-06,
+      "loss": 1.6586,
+      "step": 269
+    },
+    {
+      "epoch": 0.2773497688751926,
+      "grad_norm": 2.5443379878997803,
+      "learning_rate": 4.349315068493151e-06,
+      "loss": 1.6318,
+      "step": 270
+    },
+    {
+      "epoch": 0.2773497688751926,
+      "eval_loss": 1.9062609672546387,
+      "eval_runtime": 37.92,
+      "eval_samples_per_second": 26.371,
+      "eval_steps_per_second": 1.108,
+      "step": 270
+    },
+    {
+      "epoch": 0.278376990241397,
+      "grad_norm": 2.4064481258392334,
+      "learning_rate": 4.3664383561643846e-06,
+      "loss": 1.6178,
+      "step": 271
+    },
+    {
+      "epoch": 0.2794042116076014,
+      "grad_norm": 2.681637763977051,
+      "learning_rate": 4.383561643835616e-06,
+      "loss": 1.7071,
+      "step": 272
+    },
+    {
+      "epoch": 0.28043143297380585,
+      "grad_norm": 2.5805184841156006,
+      "learning_rate": 4.40068493150685e-06,
+      "loss": 1.6553,
+      "step": 273
+    },
+    {
+      "epoch": 0.2814586543400103,
+      "grad_norm": 2.4169883728027344,
+      "learning_rate": 4.4178082191780825e-06,
+      "loss": 1.6249,
+      "step": 274
+    },
+    {
+      "epoch": 0.2824858757062147,
+      "grad_norm": 2.557570457458496,
+      "learning_rate": 4.434931506849315e-06,
+      "loss": 1.5932,
+      "step": 275
+    },
+    {
+      "epoch": 0.2835130970724191,
+      "grad_norm": 2.3808107376098633,
+      "learning_rate": 4.4520547945205486e-06,
+      "loss": 1.6377,
+      "step": 276
+    },
+    {
+      "epoch": 0.28454031843862354,
+      "grad_norm": 2.402404546737671,
+      "learning_rate": 4.469178082191781e-06,
+      "loss": 1.6285,
+      "step": 277
+    },
+    {
+      "epoch": 0.28556753980482796,
+      "grad_norm": 2.4676456451416016,
+      "learning_rate": 4.486301369863014e-06,
+      "loss": 1.6838,
+      "step": 278
+    },
+    {
+      "epoch": 0.28659476117103233,
+      "grad_norm": 2.659083366394043,
+      "learning_rate": 4.503424657534247e-06,
+      "loss": 1.5872,
+      "step": 279
+    },
+    {
+      "epoch": 0.28762198253723675,
+      "grad_norm": 2.326733350753784,
+      "learning_rate": 4.52054794520548e-06,
+      "loss": 1.6142,
+      "step": 280
+    },
+    {
+      "epoch": 0.28762198253723675,
+      "eval_loss": 1.904759407043457,
+      "eval_runtime": 37.9651,
+      "eval_samples_per_second": 26.34,
+      "eval_steps_per_second": 1.106,
+      "step": 280
+    },
+    {
+      "epoch": 0.2886492039034412,
+      "grad_norm": 2.299504518508911,
+      "learning_rate": 4.537671232876713e-06,
+      "loss": 1.6673,
+      "step": 281
+    },
+    {
+      "epoch": 0.2896764252696456,
+      "grad_norm": 2.4932825565338135,
+      "learning_rate": 4.554794520547945e-06,
+      "loss": 1.6586,
+      "step": 282
+    },
+    {
+      "epoch": 0.29070364663585,
+      "grad_norm": 2.2599050998687744,
+      "learning_rate": 4.571917808219179e-06,
+      "loss": 1.6568,
+      "step": 283
+    },
+    {
+      "epoch": 0.29173086800205444,
+      "grad_norm": 2.435098886489868,
+      "learning_rate": 4.589041095890411e-06,
+      "loss": 1.6113,
+      "step": 284
+    },
+    {
+      "epoch": 0.29275808936825887,
+      "grad_norm": 2.539246082305908,
+      "learning_rate": 4.606164383561644e-06,
+      "loss": 1.6306,
+      "step": 285
+    },
+    {
+      "epoch": 0.2937853107344633,
+      "grad_norm": 2.596198320388794,
+      "learning_rate": 4.6232876712328774e-06,
+      "loss": 1.6473,
+      "step": 286
+    },
+    {
+      "epoch": 0.2948125321006677,
+      "grad_norm": 2.349133014678955,
+      "learning_rate": 4.64041095890411e-06,
+      "loss": 1.6042,
+      "step": 287
+    },
+    {
+      "epoch": 0.29583975346687214,
+      "grad_norm": 2.3572661876678467,
+      "learning_rate": 4.657534246575343e-06,
+      "loss": 1.6635,
+      "step": 288
+    },
+    {
+      "epoch": 0.2968669748330765,
+      "grad_norm": 2.4210307598114014,
+      "learning_rate": 4.674657534246575e-06,
+      "loss": 1.624,
+      "step": 289
+    },
+    {
+      "epoch": 0.2978941961992809,
+      "grad_norm": 2.427403688430786,
+      "learning_rate": 4.691780821917809e-06,
+      "loss": 1.6824,
+      "step": 290
+    },
+    {
+      "epoch": 0.2978941961992809,
+      "eval_loss": 1.8888994455337524,
+      "eval_runtime": 37.9777,
+      "eval_samples_per_second": 26.331,
+      "eval_steps_per_second": 1.106,
+      "step": 290
+    },
+    {
+      "epoch": 0.29892141756548535,
+      "grad_norm": 2.534029722213745,
+      "learning_rate": 4.7089041095890414e-06,
+      "loss": 1.6456,
+      "step": 291
+    },
+    {
+      "epoch": 0.29994863893168977,
+      "grad_norm": 2.329305648803711,
+      "learning_rate": 4.726027397260274e-06,
+      "loss": 1.646,
+      "step": 292
+    },
+    {
+      "epoch": 0.3009758602978942,
+      "grad_norm": 2.561548948287964,
+      "learning_rate": 4.7431506849315075e-06,
+      "loss": 1.6511,
+      "step": 293
+    },
+    {
+      "epoch": 0.3020030816640986,
+      "grad_norm": 2.707712411880493,
+      "learning_rate": 4.76027397260274e-06,
+      "loss": 1.5984,
+      "step": 294
+    },
+    {
+      "epoch": 0.30303030303030304,
+      "grad_norm": 2.403089761734009,
+      "learning_rate": 4.777397260273973e-06,
+      "loss": 1.6164,
+      "step": 295
+    },
+    {
+      "epoch": 0.30405752439650746,
+      "grad_norm": 2.424910306930542,
+      "learning_rate": 4.7945205479452054e-06,
+      "loss": 1.6291,
+      "step": 296
+    },
+    {
+      "epoch": 0.3050847457627119,
+      "grad_norm": 2.3474302291870117,
+      "learning_rate": 4.811643835616439e-06,
+      "loss": 1.553,
+      "step": 297
+    },
+    {
+      "epoch": 0.3061119671289163,
+      "grad_norm": 2.2211227416992188,
+      "learning_rate": 4.8287671232876716e-06,
+      "loss": 1.6108,
+      "step": 298
+    },
+    {
+      "epoch": 0.3071391884951207,
+      "grad_norm": 2.495983600616455,
+      "learning_rate": 4.845890410958904e-06,
+      "loss": 1.6135,
+      "step": 299
+    },
+    {
+      "epoch": 0.3081664098613251,
+      "grad_norm": 2.525495767593384,
+      "learning_rate": 4.863013698630138e-06,
+      "loss": 1.6061,
+      "step": 300
+    },
+    {
+      "epoch": 0.3081664098613251,
+      "eval_loss": 1.8982168436050415,
+      "eval_runtime": 37.9848,
+      "eval_samples_per_second": 26.326,
+      "eval_steps_per_second": 1.106,
+      "step": 300
+    },
+    {
+      "epoch": 0.3091936312275295,
+      "grad_norm": 2.449064254760742,
+      "learning_rate": 4.88013698630137e-06,
+      "loss": 1.5827,
+      "step": 301
+    },
+    {
+      "epoch": 0.31022085259373394,
+      "grad_norm": 2.288095712661743,
+      "learning_rate": 4.897260273972603e-06,
+      "loss": 1.704,
+      "step": 302
+    },
+    {
+      "epoch": 0.31124807395993837,
+      "grad_norm": 2.5301458835601807,
+      "learning_rate": 4.914383561643836e-06,
+      "loss": 1.6342,
+      "step": 303
+    },
+    {
+      "epoch": 0.3122752953261428,
+      "grad_norm": 2.2793986797332764,
+      "learning_rate": 4.931506849315069e-06,
+      "loss": 1.6688,
+      "step": 304
+    },
+    {
+      "epoch": 0.3133025166923472,
+      "grad_norm": 2.38783597946167,
+      "learning_rate": 4.948630136986302e-06,
+      "loss": 1.7014,
+      "step": 305
+    },
+    {
+      "epoch": 0.31432973805855163,
+      "grad_norm": 2.5224838256835938,
+      "learning_rate": 4.965753424657534e-06,
+      "loss": 1.5812,
+      "step": 306
+    },
+    {
+      "epoch": 0.31535695942475606,
+      "grad_norm": 2.434432029724121,
+      "learning_rate": 4.982876712328768e-06,
+      "loss": 1.6186,
+      "step": 307
+    },
+    {
+      "epoch": 0.3163841807909605,
+      "grad_norm": 2.3708460330963135,
+      "learning_rate": 5e-06,
+      "loss": 1.5669,
+      "step": 308
+    },
+    {
+      "epoch": 0.31741140215716485,
+      "grad_norm": 2.5524990558624268,
+      "learning_rate": 4.998096688237534e-06,
+      "loss": 1.6688,
+      "step": 309
+    },
+    {
+      "epoch": 0.31843862352336927,
+      "grad_norm": 2.322441577911377,
+      "learning_rate": 4.996193376475067e-06,
+      "loss": 1.5571,
+      "step": 310
+    },
+    {
+      "epoch": 0.31843862352336927,
+      "eval_loss": 1.8914265632629395,
+      "eval_runtime": 37.6505,
+      "eval_samples_per_second": 26.56,
+      "eval_steps_per_second": 1.116,
+      "step": 310
+    },
+    {
+      "epoch": 0.3194658448895737,
+      "grad_norm": 2.5248687267303467,
+      "learning_rate": 4.9942900647126005e-06,
+      "loss": 1.5585,
+      "step": 311
+    },
+    {
+      "epoch": 0.3204930662557781,
+      "grad_norm": 2.2464914321899414,
+      "learning_rate": 4.992386752950134e-06,
+      "loss": 1.5921,
+      "step": 312
+    },
+    {
+      "epoch": 0.32152028762198254,
+      "grad_norm": 2.5355331897735596,
+      "learning_rate": 4.990483441187667e-06,
+      "loss": 1.6263,
+      "step": 313
+    },
+    {
+      "epoch": 0.32254750898818696,
+      "grad_norm": 2.5611143112182617,
+      "learning_rate": 4.9885801294252e-06,
+      "loss": 1.6267,
+      "step": 314
+    },
+    {
+      "epoch": 0.3235747303543914,
+      "grad_norm": 2.364356517791748,
+      "learning_rate": 4.986676817662733e-06,
+      "loss": 1.5847,
+      "step": 315
+    },
+    {
+      "epoch": 0.3246019517205958,
+      "grad_norm": 2.455165147781372,
+      "learning_rate": 4.9847735059002665e-06,
+      "loss": 1.661,
+      "step": 316
+    },
+    {
+      "epoch": 0.3256291730868002,
+      "grad_norm": 2.4312195777893066,
+      "learning_rate": 4.9828701941378e-06,
+      "loss": 1.5995,
+      "step": 317
+    },
+    {
+      "epoch": 0.3266563944530046,
+      "grad_norm": 2.4081499576568604,
+      "learning_rate": 4.980966882375333e-06,
+      "loss": 1.5756,
+      "step": 318
+    },
+    {
+      "epoch": 0.327683615819209,
+      "grad_norm": 2.2608511447906494,
+      "learning_rate": 4.979063570612867e-06,
+      "loss": 1.543,
+      "step": 319
+    },
+    {
+      "epoch": 0.32871083718541344,
+      "grad_norm": 2.6551144123077393,
+      "learning_rate": 4.9771602588504e-06,
+      "loss": 1.6467,
+      "step": 320
+    },
+    {
+      "epoch": 0.32871083718541344,
+      "eval_loss": 1.89058518409729,
+      "eval_runtime": 37.8192,
+      "eval_samples_per_second": 26.442,
+      "eval_steps_per_second": 1.111,
+      "step": 320
+    },
+    {
+      "epoch": 0.32973805855161786,
+      "grad_norm": 2.3897273540496826,
+      "learning_rate": 4.975256947087933e-06,
+      "loss": 1.5607,
+      "step": 321
+    },
+    {
+      "epoch": 0.3307652799178223,
+      "grad_norm": 2.3869469165802,
+      "learning_rate": 4.973353635325467e-06,
+      "loss": 1.6079,
+      "step": 322
+    },
+    {
+      "epoch": 0.3317925012840267,
+      "grad_norm": 2.159193515777588,
+      "learning_rate": 4.971450323563e-06,
+      "loss": 1.5475,
+      "step": 323
+    },
+    {
+      "epoch": 0.33281972265023113,
+      "grad_norm": 2.3069653511047363,
+      "learning_rate": 4.969547011800533e-06,
+      "loss": 1.616,
+      "step": 324
+    },
+    {
+      "epoch": 0.33384694401643555,
+      "grad_norm": 2.3131422996520996,
+      "learning_rate": 4.967643700038067e-06,
+      "loss": 1.6106,
+      "step": 325
+    },
+    {
+      "epoch": 0.33487416538264,
+      "grad_norm": 2.4398200511932373,
+      "learning_rate": 4.9657403882756e-06,
+      "loss": 1.5779,
+      "step": 326
+    },
+    {
+      "epoch": 0.3359013867488444,
+      "grad_norm": 2.278024435043335,
+      "learning_rate": 4.9638370765131335e-06,
+      "loss": 1.5724,
+      "step": 327
+    },
+    {
+      "epoch": 0.33692860811504877,
+      "grad_norm": 2.4989173412323,
+      "learning_rate": 4.961933764750667e-06,
+      "loss": 1.6337,
+      "step": 328
+    },
+    {
+      "epoch": 0.3379558294812532,
+      "grad_norm": 2.46319580078125,
+      "learning_rate": 4.9600304529882e-06,
+      "loss": 1.6519,
+      "step": 329
+    },
+    {
+      "epoch": 0.3389830508474576,
+      "grad_norm": 2.2703049182891846,
+      "learning_rate": 4.9581271412257335e-06,
+      "loss": 1.6081,
+      "step": 330
+    },
+    {
+      "epoch": 0.3389830508474576,
+      "eval_loss": 1.8755545616149902,
+      "eval_runtime": 37.9058,
+      "eval_samples_per_second": 26.381,
+      "eval_steps_per_second": 1.108,
+      "step": 330
+    },
+    {
+      "epoch": 0.34001027221366203,
+      "grad_norm": 2.502196788787842,
+      "learning_rate": 4.956223829463266e-06,
+      "loss": 1.5221,
+      "step": 331
+    },
+    {
+      "epoch": 0.34103749357986646,
+      "grad_norm": 2.284280776977539,
+      "learning_rate": 4.954320517700799e-06,
+      "loss": 1.6268,
+      "step": 332
+    },
+    {
+      "epoch": 0.3420647149460709,
+      "grad_norm": 2.3551881313323975,
+      "learning_rate": 4.952417205938333e-06,
+      "loss": 1.6078,
+      "step": 333
+    },
+    {
+      "epoch": 0.3430919363122753,
+      "grad_norm": 2.343780755996704,
+      "learning_rate": 4.950513894175866e-06,
+      "loss": 1.591,
+      "step": 334
+    },
+    {
+      "epoch": 0.3441191576784797,
+      "grad_norm": 2.3524727821350098,
+      "learning_rate": 4.9486105824133995e-06,
+      "loss": 1.605,
+      "step": 335
+    },
+    {
+      "epoch": 0.34514637904468415,
+      "grad_norm": 2.349896192550659,
+      "learning_rate": 4.946707270650933e-06,
+      "loss": 1.6106,
+      "step": 336
+    },
+    {
+      "epoch": 0.34617360041088857,
+      "grad_norm": 2.5154407024383545,
+      "learning_rate": 4.944803958888466e-06,
+      "loss": 1.6429,
+      "step": 337
+    },
+    {
+      "epoch": 0.34720082177709294,
+      "grad_norm": 2.2692947387695312,
+      "learning_rate": 4.942900647126e-06,
+      "loss": 1.5865,
+      "step": 338
+    },
+    {
+      "epoch": 0.34822804314329736,
+      "grad_norm": 2.2672805786132812,
+      "learning_rate": 4.940997335363533e-06,
+      "loss": 1.5464,
+      "step": 339
+    },
+    {
+      "epoch": 0.3492552645095018,
+      "grad_norm": 2.5147039890289307,
+      "learning_rate": 4.939094023601066e-06,
+      "loss": 1.5837,
+      "step": 340
+    },
+    {
+      "epoch": 0.3492552645095018,
+      "eval_loss": 1.8672455549240112,
+      "eval_runtime": 37.9644,
+      "eval_samples_per_second": 26.34,
+      "eval_steps_per_second": 1.106,
+      "step": 340
+    },
+    {
+      "epoch": 0.3502824858757062,
+      "grad_norm": 2.3416433334350586,
+      "learning_rate": 4.9371907118386e-06,
+      "loss": 1.5937,
+      "step": 341
+    },
+    {
+      "epoch": 0.35130970724191063,
+      "grad_norm": 2.3044960498809814,
+      "learning_rate": 4.935287400076133e-06,
+      "loss": 1.6132,
+      "step": 342
+    },
+    {
+      "epoch": 0.35233692860811505,
+      "grad_norm": 2.4360430240631104,
+      "learning_rate": 4.933384088313666e-06,
+      "loss": 1.5731,
+      "step": 343
+    },
+    {
+      "epoch": 0.3533641499743195,
+      "grad_norm": 2.392050266265869,
+      "learning_rate": 4.9314807765512e-06,
+      "loss": 1.5098,
+      "step": 344
+    },
+    {
+      "epoch": 0.3543913713405239,
+      "grad_norm": 2.3815808296203613,
+      "learning_rate": 4.929577464788733e-06,
+      "loss": 1.5827,
+      "step": 345
+    },
+    {
+      "epoch": 0.3554185927067283,
+      "grad_norm": 2.423077344894409,
+      "learning_rate": 4.9276741530262665e-06,
+      "loss": 1.5857,
+      "step": 346
+    },
+    {
+      "epoch": 0.35644581407293274,
+      "grad_norm": 2.5535898208618164,
+      "learning_rate": 4.9257708412638e-06,
+      "loss": 1.5884,
+      "step": 347
+    },
+    {
+      "epoch": 0.3574730354391371,
+      "grad_norm": 2.2665445804595947,
+      "learning_rate": 4.923867529501332e-06,
+      "loss": 1.5561,
+      "step": 348
+    },
+    {
+      "epoch": 0.35850025680534153,
+      "grad_norm": 2.3357481956481934,
+      "learning_rate": 4.921964217738866e-06,
+      "loss": 1.6444,
+      "step": 349
+    },
+    {
+      "epoch": 0.35952747817154596,
+      "grad_norm": 2.4537885189056396,
+      "learning_rate": 4.920060905976399e-06,
+      "loss": 1.6262,
+      "step": 350
+    },
+    {
+      "epoch": 0.35952747817154596,
+      "eval_loss": 1.8621584177017212,
+      "eval_runtime": 38.0027,
+      "eval_samples_per_second": 26.314,
+      "eval_steps_per_second": 1.105,
+      "step": 350
+    },
+    {
+      "epoch": 0.3605546995377504,
+      "grad_norm": 2.379754066467285,
+      "learning_rate": 4.9181575942139325e-06,
+      "loss": 1.6227,
+      "step": 351
+    },
+    {
+      "epoch": 0.3615819209039548,
+      "grad_norm": 2.432905435562134,
+      "learning_rate": 4.916254282451466e-06,
+      "loss": 1.6554,
+      "step": 352
+    },
+    {
+      "epoch": 0.3626091422701592,
+      "grad_norm": 2.34236216545105,
+      "learning_rate": 4.914350970688999e-06,
+      "loss": 1.5613,
+      "step": 353
+    },
+    {
+      "epoch": 0.36363636363636365,
+      "grad_norm": 2.3019444942474365,
+      "learning_rate": 4.9124476589265325e-06,
+      "loss": 1.5535,
+      "step": 354
+    },
+    {
+      "epoch": 0.36466358500256807,
+      "grad_norm": 2.3354644775390625,
+      "learning_rate": 4.910544347164066e-06,
+      "loss": 1.6384,
+      "step": 355
+    },
+    {
+      "epoch": 0.3656908063687725,
+      "grad_norm": 2.5154788494110107,
+      "learning_rate": 4.908641035401599e-06,
+      "loss": 1.6986,
+      "step": 356
+    },
+    {
+      "epoch": 0.3667180277349769,
+      "grad_norm": 2.2707178592681885,
+      "learning_rate": 4.906737723639133e-06,
+      "loss": 1.5705,
+      "step": 357
+    },
+    {
+      "epoch": 0.3677452491011813,
+      "grad_norm": 2.3058297634124756,
+      "learning_rate": 4.904834411876665e-06,
+      "loss": 1.6033,
+      "step": 358
+    },
+    {
+      "epoch": 0.3687724704673857,
+      "grad_norm": 2.3923370838165283,
+      "learning_rate": 4.902931100114199e-06,
+      "loss": 1.5333,
+      "step": 359
+    },
+    {
+      "epoch": 0.3697996918335901,
+      "grad_norm": 2.4023196697235107,
+      "learning_rate": 4.901027788351733e-06,
+      "loss": 1.5775,
+      "step": 360
+    },
+    {
+      "epoch": 0.3697996918335901,
+      "eval_loss": 1.867856740951538,
+      "eval_runtime": 38.0125,
+      "eval_samples_per_second": 26.307,
+      "eval_steps_per_second": 1.105,
+      "step": 360
+    },
+    {
+      "epoch": 0.37082691319979455,
+      "grad_norm": 2.4903581142425537,
+      "learning_rate": 4.899124476589266e-06,
+      "loss": 1.5491,
+      "step": 361
+    },
+    {
+      "epoch": 0.371854134565999,
+      "grad_norm": 2.556683301925659,
+      "learning_rate": 4.8972211648267994e-06,
+      "loss": 1.5967,
+      "step": 362
+    },
+    {
+      "epoch": 0.3728813559322034,
+      "grad_norm": 2.3017377853393555,
+      "learning_rate": 4.895317853064333e-06,
+      "loss": 1.5661,
+      "step": 363
+    },
+    {
+      "epoch": 0.3739085772984078,
+      "grad_norm": 2.286191940307617,
+      "learning_rate": 4.893414541301866e-06,
+      "loss": 1.5679,
+      "step": 364
+    },
+    {
+      "epoch": 0.37493579866461224,
+      "grad_norm": 2.260812520980835,
+      "learning_rate": 4.891511229539399e-06,
+      "loss": 1.5965,
+      "step": 365
+    },
+    {
+      "epoch": 0.37596302003081666,
+      "grad_norm": 2.4569408893585205,
+      "learning_rate": 4.889607917776932e-06,
+      "loss": 1.5923,
+      "step": 366
+    },
+    {
+      "epoch": 0.37699024139702103,
+      "grad_norm": 2.571552038192749,
+      "learning_rate": 4.887704606014465e-06,
+      "loss": 1.6021,
+      "step": 367
+    },
+    {
+      "epoch": 0.37801746276322545,
+      "grad_norm": 2.287754774093628,
+      "learning_rate": 4.885801294251999e-06,
+      "loss": 1.5304,
+      "step": 368
+    },
+    {
+      "epoch": 0.3790446841294299,
+      "grad_norm": 2.317784070968628,
+      "learning_rate": 4.883897982489532e-06,
+      "loss": 1.6233,
+      "step": 369
+    },
+    {
+      "epoch": 0.3800719054956343,
+      "grad_norm": 2.290601968765259,
+      "learning_rate": 4.8819946707270655e-06,
+      "loss": 1.624,
+      "step": 370
+    },
+    {
+      "epoch": 0.3800719054956343,
+      "eval_loss": 1.8604528903961182,
+      "eval_runtime": 38.0176,
+      "eval_samples_per_second": 26.304,
+      "eval_steps_per_second": 1.105,
+      "step": 370
+    },
+    {
+      "epoch": 0.3810991268618387,
+      "grad_norm": 2.28519868850708,
+      "learning_rate": 4.880091358964599e-06,
+      "loss": 1.5286,
+      "step": 371
+    },
+    {
+      "epoch": 0.38212634822804314,
+      "grad_norm": 2.271810293197632,
+      "learning_rate": 4.878188047202132e-06,
+      "loss": 1.6879,
+      "step": 372
+    },
+    {
+      "epoch": 0.38315356959424757,
+      "grad_norm": 2.391848564147949,
+      "learning_rate": 4.876284735439666e-06,
+      "loss": 1.6458,
+      "step": 373
+    },
+    {
+      "epoch": 0.384180790960452,
+      "grad_norm": 2.3787031173706055,
+      "learning_rate": 4.874381423677199e-06,
+      "loss": 1.5738,
+      "step": 374
+    },
+    {
+      "epoch": 0.3852080123266564,
+      "grad_norm": 2.404219150543213,
+      "learning_rate": 4.8724781119147315e-06,
+      "loss": 1.6184,
+      "step": 375
+    },
+    {
+      "epoch": 0.38623523369286084,
+      "grad_norm": 2.372661590576172,
+      "learning_rate": 4.870574800152265e-06,
+      "loss": 1.5316,
+      "step": 376
+    },
+    {
+      "epoch": 0.3872624550590652,
+      "grad_norm": 2.4397103786468506,
+      "learning_rate": 4.868671488389798e-06,
+      "loss": 1.5605,
+      "step": 377
+    },
+    {
+      "epoch": 0.3882896764252696,
+      "grad_norm": 2.371006727218628,
+      "learning_rate": 4.8667681766273316e-06,
+      "loss": 1.5184,
+      "step": 378
+    },
+    {
+      "epoch": 0.38931689779147405,
+      "grad_norm": 2.2621910572052,
+      "learning_rate": 4.864864864864866e-06,
+      "loss": 1.5788,
+      "step": 379
+    },
+    {
+      "epoch": 0.39034411915767847,
+      "grad_norm": 2.2888615131378174,
+      "learning_rate": 4.862961553102399e-06,
+      "loss": 1.6244,
+      "step": 380
+    },
+    {
+      "epoch": 0.39034411915767847,
+      "eval_loss": 1.8586912155151367,
+      "eval_runtime": 37.9582,
+      "eval_samples_per_second": 26.345,
+      "eval_steps_per_second": 1.106,
+      "step": 380
+    },
+    {
+      "epoch": 0.3913713405238829,
+      "grad_norm": 2.4439220428466797,
+      "learning_rate": 4.8610582413399325e-06,
+      "loss": 1.5749,
+      "step": 381
+    },
+    {
+      "epoch": 0.3923985618900873,
+      "grad_norm": 2.2423763275146484,
+      "learning_rate": 4.859154929577465e-06,
+      "loss": 1.5738,
+      "step": 382
+    },
+    {
+      "epoch": 0.39342578325629174,
+      "grad_norm": 2.267467737197876,
+      "learning_rate": 4.857251617814998e-06,
+      "loss": 1.6212,
+      "step": 383
+    },
+    {
+      "epoch": 0.39445300462249616,
+      "grad_norm": 2.3680620193481445,
+      "learning_rate": 4.855348306052532e-06,
+      "loss": 1.6129,
+      "step": 384
+    },
+    {
+      "epoch": 0.3954802259887006,
+      "grad_norm": 2.3782172203063965,
+      "learning_rate": 4.853444994290065e-06,
+      "loss": 1.5605,
+      "step": 385
+    },
+    {
+      "epoch": 0.396507447354905,
+      "grad_norm": 2.5745720863342285,
+      "learning_rate": 4.8515416825275985e-06,
+      "loss": 1.634,
+      "step": 386
+    },
+    {
+      "epoch": 0.3975346687211094,
+      "grad_norm": 2.38212513923645,
+      "learning_rate": 4.849638370765132e-06,
+      "loss": 1.5671,
+      "step": 387
+    },
+    {
+      "epoch": 0.3985618900873138,
+      "grad_norm": 2.3395450115203857,
+      "learning_rate": 4.847735059002665e-06,
+      "loss": 1.5088,
+      "step": 388
+    },
+    {
+      "epoch": 0.3995891114535182,
+      "grad_norm": 2.474440574645996,
+      "learning_rate": 4.8458317472401985e-06,
+      "loss": 1.5997,
+      "step": 389
+    },
+    {
+      "epoch": 0.40061633281972264,
+      "grad_norm": 2.429133176803589,
+      "learning_rate": 4.843928435477732e-06,
+      "loss": 1.5611,
+      "step": 390
+    },
+    {
+      "epoch": 0.40061633281972264,
+      "eval_loss": 1.8613603115081787,
+      "eval_runtime": 37.9862,
+      "eval_samples_per_second": 26.325,
+      "eval_steps_per_second": 1.106,
+      "step": 390
+    },
+    {
+      "epoch": 0.40164355418592707,
+      "grad_norm": 2.403207778930664,
+      "learning_rate": 4.842025123715265e-06,
+      "loss": 1.5976,
+      "step": 391
+    },
+    {
+      "epoch": 0.4026707755521315,
+      "grad_norm": 2.383018732070923,
+      "learning_rate": 4.840121811952798e-06,
+      "loss": 1.5791,
+      "step": 392
+    },
+    {
+      "epoch": 0.4036979969183359,
+      "grad_norm": 2.5303306579589844,
+      "learning_rate": 4.838218500190331e-06,
+      "loss": 1.6035,
+      "step": 393
+    },
+    {
+      "epoch": 0.40472521828454033,
+      "grad_norm": 2.391075372695923,
+      "learning_rate": 4.8363151884278645e-06,
+      "loss": 1.5795,
+      "step": 394
+    },
+    {
+      "epoch": 0.40575243965074476,
+      "grad_norm": 2.396934747695923,
+      "learning_rate": 4.834411876665398e-06,
+      "loss": 1.5965,
+      "step": 395
+    },
+    {
+      "epoch": 0.4067796610169492,
+      "grad_norm": 2.4631423950195312,
+      "learning_rate": 4.832508564902931e-06,
+      "loss": 1.5372,
+      "step": 396
+    },
+    {
+      "epoch": 0.40780688238315355,
+      "grad_norm": 2.3457326889038086,
+      "learning_rate": 4.830605253140465e-06,
+      "loss": 1.5729,
+      "step": 397
+    },
+    {
+      "epoch": 0.40883410374935797,
+      "grad_norm": 2.3763277530670166,
+      "learning_rate": 4.828701941377998e-06,
+      "loss": 1.5309,
+      "step": 398
+    },
+    {
+      "epoch": 0.4098613251155624,
+      "grad_norm": 2.442439556121826,
+      "learning_rate": 4.826798629615531e-06,
+      "loss": 1.5465,
+      "step": 399
+    },
+    {
+      "epoch": 0.4108885464817668,
+      "grad_norm": 2.3464581966400146,
+      "learning_rate": 4.824895317853065e-06,
+      "loss": 1.5665,
+      "step": 400
+    },
+    {
+      "epoch": 0.4108885464817668,
+      "eval_loss": 1.8671027421951294,
+      "eval_runtime": 38.0166,
+      "eval_samples_per_second": 26.304,
+      "eval_steps_per_second": 1.105,
+      "step": 400
+    },
+    {
+      "epoch": 0.41191576784797124,
+      "grad_norm": 2.5441887378692627,
+      "learning_rate": 4.822992006090598e-06,
+      "loss": 1.5232,
+      "step": 401
+    },
+    {
+      "epoch": 0.41294298921417566,
+      "grad_norm": 2.4234976768493652,
+      "learning_rate": 4.821088694328131e-06,
+      "loss": 1.5569,
+      "step": 402
+    },
+    {
+      "epoch": 0.4139702105803801,
+      "grad_norm": 2.3603928089141846,
+      "learning_rate": 4.819185382565665e-06,
+      "loss": 1.612,
+      "step": 403
+    },
+    {
+      "epoch": 0.4149974319465845,
+      "grad_norm": 2.337862491607666,
+      "learning_rate": 4.817282070803198e-06,
+      "loss": 1.5882,
+      "step": 404
+    },
+    {
+      "epoch": 0.41602465331278893,
+      "grad_norm": 2.3805742263793945,
+      "learning_rate": 4.8153787590407315e-06,
+      "loss": 1.5204,
+      "step": 405
+    },
+    {
+      "epoch": 0.41705187467899335,
+      "grad_norm": 2.3335204124450684,
+      "learning_rate": 4.813475447278265e-06,
+      "loss": 1.6041,
+      "step": 406
+    },
+    {
+      "epoch": 0.4180790960451977,
+      "grad_norm": 2.406137466430664,
+      "learning_rate": 4.811572135515798e-06,
+      "loss": 1.5348,
+      "step": 407
+    },
+    {
+      "epoch": 0.41910631741140214,
+      "grad_norm": 2.2823917865753174,
+      "learning_rate": 4.809668823753332e-06,
+      "loss": 1.5351,
+      "step": 408
+    },
+    {
+      "epoch": 0.42013353877760656,
+      "grad_norm": 2.3577091693878174,
+      "learning_rate": 4.807765511990864e-06,
+      "loss": 1.5733,
+      "step": 409
+    },
+    {
+      "epoch": 0.421160760143811,
+      "grad_norm": 2.536691665649414,
+      "learning_rate": 4.8058622002283975e-06,
+      "loss": 1.6153,
+      "step": 410
+    },
+    {
+      "epoch": 0.421160760143811,
+      "eval_loss": 1.8613901138305664,
+      "eval_runtime": 37.6689,
+      "eval_samples_per_second": 26.547,
+      "eval_steps_per_second": 1.115,
+      "step": 410
+    },
+    {
+      "epoch": 0.4221879815100154,
+      "grad_norm": 2.4020118713378906,
+      "learning_rate": 4.803958888465931e-06,
+      "loss": 1.5722,
+      "step": 411
+    },
+    {
+      "epoch": 0.42321520287621983,
+      "grad_norm": 2.5284810066223145,
+      "learning_rate": 4.802055576703464e-06,
+      "loss": 1.6546,
+      "step": 412
+    },
+    {
+      "epoch": 0.42424242424242425,
+      "grad_norm": 2.270094156265259,
+      "learning_rate": 4.8001522649409976e-06,
+      "loss": 1.5345,
+      "step": 413
+    },
+    {
+      "epoch": 0.4252696456086287,
+      "grad_norm": 2.241417407989502,
+      "learning_rate": 4.798248953178531e-06,
+      "loss": 1.5167,
+      "step": 414
+    },
+    {
+      "epoch": 0.4262968669748331,
+      "grad_norm": 2.3733017444610596,
+      "learning_rate": 4.796345641416064e-06,
+      "loss": 1.5532,
+      "step": 415
+    },
+    {
+      "epoch": 0.42732408834103747,
+      "grad_norm": 2.452587842941284,
+      "learning_rate": 4.794442329653598e-06,
+      "loss": 1.6136,
+      "step": 416
+    },
+    {
+      "epoch": 0.4283513097072419,
+      "grad_norm": 2.417217969894409,
+      "learning_rate": 4.792539017891131e-06,
+      "loss": 1.5398,
+      "step": 417
+    },
+    {
+      "epoch": 0.4293785310734463,
+      "grad_norm": 2.4502453804016113,
+      "learning_rate": 4.790635706128664e-06,
+      "loss": 1.5741,
+      "step": 418
+    },
+    {
+      "epoch": 0.43040575243965074,
+      "grad_norm": 2.2996811866760254,
+      "learning_rate": 4.788732394366197e-06,
+      "loss": 1.511,
+      "step": 419
+    },
+    {
+      "epoch": 0.43143297380585516,
+      "grad_norm": 2.3573555946350098,
+      "learning_rate": 4.786829082603731e-06,
+      "loss": 1.5117,
+      "step": 420
+    },
+    {
+      "epoch": 0.43143297380585516,
+      "eval_loss": 1.8690649271011353,
+      "eval_runtime": 37.7813,
+      "eval_samples_per_second": 26.468,
+      "eval_steps_per_second": 1.112,
+      "step": 420
+    },
+    {
+      "epoch": 0.4324601951720596,
+      "grad_norm": 2.37233567237854,
+      "learning_rate": 4.7849257708412645e-06,
+      "loss": 1.5873,
+      "step": 421
+    },
+    {
+      "epoch": 0.433487416538264,
+      "grad_norm": 2.146909713745117,
+      "learning_rate": 4.783022459078798e-06,
+      "loss": 1.5291,
+      "step": 422
+    },
+    {
+      "epoch": 0.4345146379044684,
+      "grad_norm": 2.5237205028533936,
+      "learning_rate": 4.781119147316331e-06,
+      "loss": 1.5545,
+      "step": 423
+    },
+    {
+      "epoch": 0.43554185927067285,
+      "grad_norm": 2.356842279434204,
+      "learning_rate": 4.7792158355538645e-06,
+      "loss": 1.5635,
+      "step": 424
+    },
+    {
+      "epoch": 0.43656908063687727,
+      "grad_norm": 2.1963610649108887,
+      "learning_rate": 4.777312523791398e-06,
+      "loss": 1.5355,
+      "step": 425
+    },
+    {
+      "epoch": 0.43759630200308164,
+      "grad_norm": 2.6425886154174805,
+      "learning_rate": 4.77540921202893e-06,
+      "loss": 1.5803,
+      "step": 426
+    },
+    {
+      "epoch": 0.43862352336928606,
+      "grad_norm": 2.4072272777557373,
+      "learning_rate": 4.773505900266464e-06,
+      "loss": 1.5649,
+      "step": 427
+    },
+    {
+      "epoch": 0.4396507447354905,
+      "grad_norm": 2.592374801635742,
+      "learning_rate": 4.771602588503997e-06,
+      "loss": 1.5693,
+      "step": 428
+    },
+    {
+      "epoch": 0.4406779661016949,
+      "grad_norm": 2.4534225463867188,
+      "learning_rate": 4.7696992767415305e-06,
+      "loss": 1.591,
+      "step": 429
+    },
+    {
+      "epoch": 0.44170518746789933,
+      "grad_norm": 2.434457778930664,
+      "learning_rate": 4.767795964979064e-06,
+      "loss": 1.5991,
+      "step": 430
+    },
+    {
+      "epoch": 0.44170518746789933,
+      "eval_loss": 1.8584965467453003,
+      "eval_runtime": 37.8854,
+      "eval_samples_per_second": 26.395,
+      "eval_steps_per_second": 1.109,
+      "step": 430
+    },
+    {
+      "epoch": 0.44273240883410375,
+      "grad_norm": 2.3150746822357178,
+      "learning_rate": 4.765892653216597e-06,
+      "loss": 1.6032,
+      "step": 431
+    },
+    {
+      "epoch": 0.4437596302003082,
+      "grad_norm": 2.3040223121643066,
+      "learning_rate": 4.763989341454131e-06,
+      "loss": 1.5115,
+      "step": 432
+    },
+    {
+      "epoch": 0.4447868515665126,
+      "grad_norm": 2.413007974624634,
+      "learning_rate": 4.762086029691664e-06,
+      "loss": 1.5305,
+      "step": 433
+    },
+    {
+      "epoch": 0.445814072932717,
+      "grad_norm": 2.526289463043213,
+      "learning_rate": 4.760182717929197e-06,
+      "loss": 1.564,
+      "step": 434
+    },
+    {
+      "epoch": 0.44684129429892144,
+      "grad_norm": 2.4122776985168457,
+      "learning_rate": 4.758279406166731e-06,
+      "loss": 1.4918,
+      "step": 435
+    },
+    {
+      "epoch": 0.4478685156651258,
+      "grad_norm": 2.474274158477783,
+      "learning_rate": 4.756376094404263e-06,
+      "loss": 1.5583,
+      "step": 436
+    },
+    {
+      "epoch": 0.44889573703133023,
+      "grad_norm": 2.5322892665863037,
+      "learning_rate": 4.7544727826417966e-06,
+      "loss": 1.6081,
+      "step": 437
+    },
+    {
+      "epoch": 0.44992295839753466,
+      "grad_norm": 2.391660690307617,
+      "learning_rate": 4.75256947087933e-06,
+      "loss": 1.5628,
+      "step": 438
+    },
+    {
+      "epoch": 0.4509501797637391,
+      "grad_norm": 2.426671028137207,
+      "learning_rate": 4.750666159116863e-06,
+      "loss": 1.531,
+      "step": 439
+    },
+    {
+      "epoch": 0.4519774011299435,
+      "grad_norm": 2.7095694541931152,
+      "learning_rate": 4.7487628473543975e-06,
+      "loss": 1.4811,
+      "step": 440
+    },
+    {
+      "epoch": 0.4519774011299435,
+      "eval_loss": 1.8348888158798218,
+      "eval_runtime": 37.9114,
+      "eval_samples_per_second": 26.377,
+      "eval_steps_per_second": 1.108,
+      "step": 440
+    },
+    {
+      "epoch": 0.4530046224961479,
+      "grad_norm": 2.5425240993499756,
+      "learning_rate": 4.746859535591931e-06,
+      "loss": 1.5453,
+      "step": 441
+    },
+    {
+      "epoch": 0.45403184386235235,
+      "grad_norm": 2.6848669052124023,
+      "learning_rate": 4.744956223829464e-06,
+      "loss": 1.526,
+      "step": 442
+    },
+    {
+      "epoch": 0.45505906522855677,
+      "grad_norm": 2.634068489074707,
+      "learning_rate": 4.743052912066997e-06,
+      "loss": 1.5748,
+      "step": 443
+    },
+    {
+      "epoch": 0.4560862865947612,
+      "grad_norm": 2.492030382156372,
+      "learning_rate": 4.74114960030453e-06,
+      "loss": 1.5858,
+      "step": 444
+    },
+    {
+      "epoch": 0.4571135079609656,
+      "grad_norm": 2.787740707397461,
+      "learning_rate": 4.7392462885420635e-06,
+      "loss": 1.4774,
+      "step": 445
+    },
+    {
+      "epoch": 0.45814072932717,
+      "grad_norm": 2.395914077758789,
+      "learning_rate": 4.737342976779597e-06,
+      "loss": 1.558,
+      "step": 446
+    },
+    {
+      "epoch": 0.4591679506933744,
+      "grad_norm": 2.6642563343048096,
+      "learning_rate": 4.73543966501713e-06,
+      "loss": 1.5558,
+      "step": 447
+    },
+    {
+      "epoch": 0.4601951720595788,
+      "grad_norm": 2.619943857192993,
+      "learning_rate": 4.7335363532546636e-06,
+      "loss": 1.5442,
+      "step": 448
+    },
+    {
+      "epoch": 0.46122239342578325,
+      "grad_norm": 2.355133056640625,
+      "learning_rate": 4.731633041492197e-06,
+      "loss": 1.5613,
+      "step": 449
+    },
+    {
+      "epoch": 0.4622496147919877,
+      "grad_norm": 2.6572225093841553,
+      "learning_rate": 4.72972972972973e-06,
+      "loss": 1.5951,
+      "step": 450
+    },
+    {
+      "epoch": 0.4622496147919877,
+      "eval_loss": 1.83199143409729,
+      "eval_runtime": 37.8996,
+      "eval_samples_per_second": 26.385,
+      "eval_steps_per_second": 1.108,
+      "step": 450
+    },
+    {
+      "epoch": 0.4632768361581921,
+      "grad_norm": 2.692873001098633,
+      "learning_rate": 4.727826417967264e-06,
+      "loss": 1.5531,
+      "step": 451
+    },
+    {
+      "epoch": 0.4643040575243965,
+      "grad_norm": 2.3903489112854004,
+      "learning_rate": 4.725923106204797e-06,
+      "loss": 1.5411,
+      "step": 452
+    },
+    {
+      "epoch": 0.46533127889060094,
+      "grad_norm": 2.453263521194458,
+      "learning_rate": 4.72401979444233e-06,
+      "loss": 1.5632,
+      "step": 453
+    },
+    {
+      "epoch": 0.46635850025680536,
+      "grad_norm": 2.5637760162353516,
+      "learning_rate": 4.722116482679863e-06,
+      "loss": 1.5629,
+      "step": 454
+    },
+    {
+      "epoch": 0.46738572162300973,
+      "grad_norm": 2.387033700942993,
+      "learning_rate": 4.720213170917396e-06,
+      "loss": 1.5786,
+      "step": 455
+    },
+    {
+      "epoch": 0.46841294298921415,
+      "grad_norm": 2.4563071727752686,
+      "learning_rate": 4.71830985915493e-06,
+      "loss": 1.5321,
+      "step": 456
+    },
+    {
+      "epoch": 0.4694401643554186,
+      "grad_norm": 2.607451915740967,
+      "learning_rate": 4.716406547392463e-06,
+      "loss": 1.5958,
+      "step": 457
+    },
+    {
+      "epoch": 0.470467385721623,
+      "grad_norm": 2.4846854209899902,
+      "learning_rate": 4.714503235629996e-06,
+      "loss": 1.5205,
+      "step": 458
+    },
+    {
+      "epoch": 0.4714946070878274,
+      "grad_norm": 2.377629041671753,
+      "learning_rate": 4.71259992386753e-06,
+      "loss": 1.5403,
+      "step": 459
+    },
+    {
+      "epoch": 0.47252182845403184,
+      "grad_norm": 2.4470160007476807,
+      "learning_rate": 4.710696612105063e-06,
+      "loss": 1.5456,
+      "step": 460
+    },
+    {
+      "epoch": 0.47252182845403184,
+      "eval_loss": 1.83836829662323,
+      "eval_runtime": 37.88,
+      "eval_samples_per_second": 26.399,
+      "eval_steps_per_second": 1.109,
+      "step": 460
+    },
+    {
+      "epoch": 0.47354904982023627,
+      "grad_norm": 2.383418560028076,
+      "learning_rate": 4.708793300342596e-06,
+      "loss": 1.5468,
+      "step": 461
+    },
+    {
+      "epoch": 0.4745762711864407,
+      "grad_norm": 2.3672895431518555,
+      "learning_rate": 4.70688998858013e-06,
+      "loss": 1.5313,
+      "step": 462
+    },
+    {
+      "epoch": 0.4756034925526451,
+      "grad_norm": 2.415717124938965,
+      "learning_rate": 4.704986676817663e-06,
+      "loss": 1.5645,
+      "step": 463
+    },
+    {
+      "epoch": 0.47663071391884954,
+      "grad_norm": 2.4690067768096924,
+      "learning_rate": 4.7030833650551965e-06,
+      "loss": 1.5285,
+      "step": 464
+    },
+    {
+      "epoch": 0.4776579352850539,
+      "grad_norm": 2.3845672607421875,
+      "learning_rate": 4.70118005329273e-06,
+      "loss": 1.596,
+      "step": 465
+    },
+    {
+      "epoch": 0.4786851566512583,
+      "grad_norm": 2.2801802158355713,
+      "learning_rate": 4.699276741530263e-06,
+      "loss": 1.5777,
+      "step": 466
+    },
+    {
+      "epoch": 0.47971237801746275,
+      "grad_norm": 2.4668991565704346,
+      "learning_rate": 4.697373429767797e-06,
+      "loss": 1.5566,
+      "step": 467
+    },
+    {
+      "epoch": 0.48073959938366717,
+      "grad_norm": 2.446078062057495,
+      "learning_rate": 4.69547011800533e-06,
+      "loss": 1.5664,
+      "step": 468
+    },
+    {
+      "epoch": 0.4817668207498716,
+      "grad_norm": 2.2310681343078613,
+      "learning_rate": 4.693566806242863e-06,
+      "loss": 1.5425,
+      "step": 469
+    },
+    {
+      "epoch": 0.482794042116076,
+      "grad_norm": 2.3938229084014893,
+      "learning_rate": 4.691663494480397e-06,
+      "loss": 1.576,
+      "step": 470
+    },
+    {
+      "epoch": 0.482794042116076,
+      "eval_loss": 1.83778977394104,
+      "eval_runtime": 37.9276,
+      "eval_samples_per_second": 26.366,
+      "eval_steps_per_second": 1.107,
+      "step": 470
+    },
+    {
+      "epoch": 0.48382126348228044,
+      "grad_norm": 2.23827862739563,
+      "learning_rate": 4.689760182717929e-06,
+      "loss": 1.4763,
+      "step": 471
+    },
+    {
+      "epoch": 0.48484848484848486,
+      "grad_norm": 2.3148930072784424,
+      "learning_rate": 4.6878568709554626e-06,
+      "loss": 1.5386,
+      "step": 472
+    },
+    {
+      "epoch": 0.4858757062146893,
+      "grad_norm": 2.407036781311035,
+      "learning_rate": 4.685953559192996e-06,
+      "loss": 1.5576,
+      "step": 473
+    },
+    {
+      "epoch": 0.4869029275808937,
+      "grad_norm": 2.4675374031066895,
+      "learning_rate": 4.684050247430529e-06,
+      "loss": 1.4931,
+      "step": 474
+    },
+    {
+      "epoch": 0.4879301489470981,
+      "grad_norm": 2.4593775272369385,
+      "learning_rate": 4.682146935668063e-06,
+      "loss": 1.5348,
+      "step": 475
+    },
+    {
+      "epoch": 0.4889573703133025,
+      "grad_norm": 2.4169375896453857,
+      "learning_rate": 4.680243623905596e-06,
+      "loss": 1.5614,
+      "step": 476
+    },
+    {
+      "epoch": 0.4899845916795069,
+      "grad_norm": 2.6177303791046143,
+      "learning_rate": 4.678340312143129e-06,
+      "loss": 1.5138,
+      "step": 477
+    },
+    {
+      "epoch": 0.49101181304571134,
+      "grad_norm": 2.6418113708496094,
+      "learning_rate": 4.676437000380663e-06,
+      "loss": 1.6377,
+      "step": 478
+    },
+    {
+      "epoch": 0.49203903441191577,
+      "grad_norm": 2.338491439819336,
+      "learning_rate": 4.674533688618196e-06,
+      "loss": 1.5682,
+      "step": 479
+    },
+    {
+      "epoch": 0.4930662557781202,
+      "grad_norm": 2.5901622772216797,
+      "learning_rate": 4.6726303768557295e-06,
+      "loss": 1.4899,
+      "step": 480
+    },
+    {
+      "epoch": 0.4930662557781202,
+      "eval_loss": 1.8238886594772339,
+      "eval_runtime": 37.9605,
+      "eval_samples_per_second": 26.343,
+      "eval_steps_per_second": 1.106,
+      "step": 480
+    },
+    {
+      "epoch": 0.4940934771443246,
+      "grad_norm": 2.470827102661133,
+      "learning_rate": 4.670727065093263e-06,
+      "loss": 1.5016,
+      "step": 481
+    },
+    {
+      "epoch": 0.49512069851052903,
+      "grad_norm": 2.371201992034912,
+      "learning_rate": 4.668823753330796e-06,
+      "loss": 1.5399,
+      "step": 482
+    },
+    {
+      "epoch": 0.49614791987673346,
+      "grad_norm": 2.532802104949951,
+      "learning_rate": 4.6669204415683295e-06,
+      "loss": 1.5985,
+      "step": 483
+    },
+    {
+      "epoch": 0.4971751412429379,
+      "grad_norm": 2.6294727325439453,
+      "learning_rate": 4.665017129805863e-06,
+      "loss": 1.5802,
+      "step": 484
+    },
+    {
+      "epoch": 0.49820236260914225,
+      "grad_norm": 2.2164878845214844,
+      "learning_rate": 4.663113818043396e-06,
+      "loss": 1.5384,
+      "step": 485
+    },
+    {
+      "epoch": 0.49922958397534667,
+      "grad_norm": 2.311232805252075,
+      "learning_rate": 4.66121050628093e-06,
+      "loss": 1.5785,
+      "step": 486
+    },
+    {
+      "epoch": 0.5002568053415511,
+      "grad_norm": 2.2916736602783203,
+      "learning_rate": 4.659307194518463e-06,
+      "loss": 1.5401,
+      "step": 487
+    },
+    {
+      "epoch": 0.5012840267077555,
+      "grad_norm": 2.3978800773620605,
+      "learning_rate": 4.6574038827559955e-06,
+      "loss": 1.5619,
+      "step": 488
+    },
+    {
+      "epoch": 0.50231124807396,
+      "grad_norm": 2.438917398452759,
+      "learning_rate": 4.655500570993529e-06,
+      "loss": 1.5824,
+      "step": 489
+    },
+    {
+      "epoch": 0.5033384694401644,
+      "grad_norm": 2.528862237930298,
+      "learning_rate": 4.653597259231062e-06,
+      "loss": 1.4802,
+      "step": 490
+    },
+    {
+      "epoch": 0.5033384694401644,
+      "eval_loss": 1.83504056930542,
+      "eval_runtime": 37.9357,
+      "eval_samples_per_second": 26.36,
+      "eval_steps_per_second": 1.107,
+      "step": 490
+    },
+    {
+      "epoch": 0.5043656908063687,
+      "grad_norm": 2.54522442817688,
+      "learning_rate": 4.651693947468596e-06,
+      "loss": 1.6186,
+      "step": 491
+    },
+    {
+      "epoch": 0.5053929121725732,
+      "grad_norm": 2.5037031173706055,
+      "learning_rate": 4.649790635706129e-06,
+      "loss": 1.5159,
+      "step": 492
+    },
+    {
+      "epoch": 0.5064201335387776,
+      "grad_norm": 2.4600465297698975,
+      "learning_rate": 4.647887323943662e-06,
+      "loss": 1.5378,
+      "step": 493
+    },
+    {
+      "epoch": 0.507447354904982,
+      "grad_norm": 2.393695831298828,
+      "learning_rate": 4.645984012181196e-06,
+      "loss": 1.4485,
+      "step": 494
+    },
+    {
+      "epoch": 0.5084745762711864,
+      "grad_norm": 2.488626718521118,
+      "learning_rate": 4.644080700418729e-06,
+      "loss": 1.57,
+      "step": 495
+    },
+    {
+      "epoch": 0.5095017976373909,
+      "grad_norm": 2.409858465194702,
+      "learning_rate": 4.642177388656262e-06,
+      "loss": 1.5343,
+      "step": 496
+    },
+    {
+      "epoch": 0.5105290190035953,
+      "grad_norm": 2.5607573986053467,
+      "learning_rate": 4.640274076893796e-06,
+      "loss": 1.5825,
+      "step": 497
+    },
+    {
+      "epoch": 0.5115562403697997,
+      "grad_norm": 2.5953354835510254,
+      "learning_rate": 4.638370765131328e-06,
+      "loss": 1.5129,
+      "step": 498
+    },
+    {
+      "epoch": 0.5125834617360041,
+      "grad_norm": 2.3321897983551025,
+      "learning_rate": 4.636467453368862e-06,
+      "loss": 1.4972,
+      "step": 499
+    },
+    {
+      "epoch": 0.5136106831022085,
+      "grad_norm": 2.332052707672119,
+      "learning_rate": 4.634564141606396e-06,
+      "loss": 1.567,
+      "step": 500
+    },
+    {
+      "epoch": 0.5136106831022085,
+      "eval_loss": 1.824558138847351,
+      "eval_runtime": 37.9157,
+      "eval_samples_per_second": 26.374,
+      "eval_steps_per_second": 1.108,
+      "step": 500
+    },
+    {
+      "epoch": 0.514637904468413,
+      "grad_norm": 2.431809902191162,
+      "learning_rate": 4.632660829843929e-06,
+      "loss": 1.494,
+      "step": 501
+    },
+    {
+      "epoch": 0.5156651258346173,
+      "grad_norm": 2.7118983268737793,
+      "learning_rate": 4.630757518081463e-06,
+      "loss": 1.6289,
+      "step": 502
+    },
+    {
+      "epoch": 0.5166923472008218,
+      "grad_norm": 2.3549370765686035,
+      "learning_rate": 4.628854206318996e-06,
+      "loss": 1.5363,
+      "step": 503
+    },
+    {
+      "epoch": 0.5177195685670262,
+      "grad_norm": 2.3343000411987305,
+      "learning_rate": 4.626950894556529e-06,
+      "loss": 1.5749,
+      "step": 504
+    },
+    {
+      "epoch": 0.5187467899332306,
+      "grad_norm": 2.4281466007232666,
+      "learning_rate": 4.625047582794062e-06,
+      "loss": 1.4958,
+      "step": 505
+    },
+    {
+      "epoch": 0.519774011299435,
+      "grad_norm": 2.3858845233917236,
+      "learning_rate": 4.623144271031595e-06,
+      "loss": 1.5031,
+      "step": 506
+    },
+    {
+      "epoch": 0.5208012326656395,
+      "grad_norm": 2.3714635372161865,
+      "learning_rate": 4.6212409592691286e-06,
+      "loss": 1.5808,
+      "step": 507
+    },
+    {
+      "epoch": 0.5218284540318439,
+      "grad_norm": 2.5290520191192627,
+      "learning_rate": 4.619337647506662e-06,
+      "loss": 1.541,
+      "step": 508
+    },
+    {
+      "epoch": 0.5228556753980482,
+      "grad_norm": 2.5521368980407715,
+      "learning_rate": 4.617434335744195e-06,
+      "loss": 1.4992,
+      "step": 509
+    },
+    {
+      "epoch": 0.5238828967642527,
+      "grad_norm": 2.436588764190674,
+      "learning_rate": 4.615531023981729e-06,
+      "loss": 1.5283,
+      "step": 510
+    },
+    {
+      "epoch": 0.5238828967642527,
+      "eval_loss": 1.8220165967941284,
+      "eval_runtime": 37.6123,
+      "eval_samples_per_second": 26.587,
+      "eval_steps_per_second": 1.117,
+      "step": 510
+    },
+    {
+      "epoch": 0.5249101181304571,
+      "grad_norm": 2.5240023136138916,
+      "learning_rate": 4.613627712219262e-06,
+      "loss": 1.5426,
+      "step": 511
+    },
+    {
+      "epoch": 0.5259373394966615,
+      "grad_norm": 2.5292775630950928,
+      "learning_rate": 4.611724400456795e-06,
+      "loss": 1.5411,
+      "step": 512
+    },
+    {
+      "epoch": 0.5269645608628659,
+      "grad_norm": 2.3622210025787354,
+      "learning_rate": 4.609821088694329e-06,
+      "loss": 1.5083,
+      "step": 513
+    },
+    {
+      "epoch": 0.5279917822290704,
+      "grad_norm": 2.545267105102539,
+      "learning_rate": 4.607917776931862e-06,
+      "loss": 1.5541,
+      "step": 514
+    },
+    {
+      "epoch": 0.5290190035952748,
+      "grad_norm": 2.3417961597442627,
+      "learning_rate": 4.606014465169395e-06,
+      "loss": 1.5686,
+      "step": 515
+    },
+    {
+      "epoch": 0.5300462249614792,
+      "grad_norm": 2.2782580852508545,
+      "learning_rate": 4.604111153406928e-06,
+      "loss": 1.4761,
+      "step": 516
+    },
+    {
+      "epoch": 0.5310734463276836,
+      "grad_norm": 2.4437527656555176,
+      "learning_rate": 4.602207841644461e-06,
+      "loss": 1.5818,
+      "step": 517
+    },
+    {
+      "epoch": 0.5321006676938881,
+      "grad_norm": 2.3594319820404053,
+      "learning_rate": 4.600304529881995e-06,
+      "loss": 1.5556,
+      "step": 518
+    },
+    {
+      "epoch": 0.5331278890600925,
+      "grad_norm": 2.325979471206665,
+      "learning_rate": 4.598401218119528e-06,
+      "loss": 1.5726,
+      "step": 519
+    },
+    {
+      "epoch": 0.5341551104262968,
+      "grad_norm": 2.2765676975250244,
+      "learning_rate": 4.5964979063570614e-06,
+      "loss": 1.5151,
+      "step": 520
+    },
+    {
+      "epoch": 0.5341551104262968,
+      "eval_loss": 1.825669288635254,
+      "eval_runtime": 37.7491,
+      "eval_samples_per_second": 26.491,
+      "eval_steps_per_second": 1.113,
+      "step": 520
+    },
+    {
+      "epoch": 0.5351823317925013,
+      "grad_norm": 2.357647657394409,
+      "learning_rate": 4.594594594594596e-06,
+      "loss": 1.5505,
+      "step": 521
+    },
+    {
+      "epoch": 0.5362095531587057,
+      "grad_norm": 2.3652896881103516,
+      "learning_rate": 4.592691282832128e-06,
+      "loss": 1.5621,
+      "step": 522
+    },
+    {
+      "epoch": 0.5372367745249101,
+      "grad_norm": 2.234096050262451,
+      "learning_rate": 4.5907879710696615e-06,
+      "loss": 1.4851,
+      "step": 523
+    },
+    {
+      "epoch": 0.5382639958911145,
+      "grad_norm": 2.3414828777313232,
+      "learning_rate": 4.588884659307195e-06,
+      "loss": 1.5594,
+      "step": 524
+    },
+    {
+      "epoch": 0.539291217257319,
+      "grad_norm": 2.3934929370880127,
+      "learning_rate": 4.586981347544728e-06,
+      "loss": 1.4994,
+      "step": 525
+    },
+    {
+      "epoch": 0.5403184386235234,
+      "grad_norm": 2.400479555130005,
+      "learning_rate": 4.585078035782262e-06,
+      "loss": 1.5346,
+      "step": 526
+    },
+    {
+      "epoch": 0.5413456599897278,
+      "grad_norm": 2.3817873001098633,
+      "learning_rate": 4.583174724019795e-06,
+      "loss": 1.4803,
+      "step": 527
+    },
+    {
+      "epoch": 0.5423728813559322,
+      "grad_norm": 2.4503185749053955,
+      "learning_rate": 4.581271412257328e-06,
+      "loss": 1.4711,
+      "step": 528
+    },
+    {
+      "epoch": 0.5434001027221366,
+      "grad_norm": 2.452578067779541,
+      "learning_rate": 4.579368100494862e-06,
+      "loss": 1.5645,
+      "step": 529
+    },
+    {
+      "epoch": 0.544427324088341,
+      "grad_norm": 2.3278186321258545,
+      "learning_rate": 4.577464788732395e-06,
+      "loss": 1.5428,
+      "step": 530
+    },
+    {
+      "epoch": 0.544427324088341,
+      "eval_loss": 1.8242785930633545,
+      "eval_runtime": 37.9229,
+      "eval_samples_per_second": 26.369,
+      "eval_steps_per_second": 1.108,
+      "step": 530
+    },
+    {
+      "epoch": 0.5454545454545454,
+      "grad_norm": 2.5192365646362305,
+      "learning_rate": 4.575561476969928e-06,
+      "loss": 1.5896,
+      "step": 531
+    },
+    {
+      "epoch": 0.5464817668207499,
+      "grad_norm": 2.308701276779175,
+      "learning_rate": 4.573658165207461e-06,
+      "loss": 1.5518,
+      "step": 532
+    },
+    {
+      "epoch": 0.5475089881869543,
+      "grad_norm": 2.4221107959747314,
+      "learning_rate": 4.571754853444994e-06,
+      "loss": 1.5387,
+      "step": 533
+    },
+    {
+      "epoch": 0.5485362095531587,
+      "grad_norm": 2.457606077194214,
+      "learning_rate": 4.569851541682528e-06,
+      "loss": 1.4997,
+      "step": 534
+    },
+    {
+      "epoch": 0.5495634309193631,
+      "grad_norm": 2.3406503200531006,
+      "learning_rate": 4.567948229920061e-06,
+      "loss": 1.4761,
+      "step": 535
+    },
+    {
+      "epoch": 0.5505906522855676,
+      "grad_norm": 2.555712938308716,
+      "learning_rate": 4.566044918157594e-06,
+      "loss": 1.5114,
+      "step": 536
+    },
+    {
+      "epoch": 0.551617873651772,
+      "grad_norm": 2.373013734817505,
+      "learning_rate": 4.564141606395128e-06,
+      "loss": 1.5516,
+      "step": 537
+    },
+    {
+      "epoch": 0.5526450950179764,
+      "grad_norm": 2.3915255069732666,
+      "learning_rate": 4.562238294632661e-06,
+      "loss": 1.5145,
+      "step": 538
+    },
+    {
+      "epoch": 0.5536723163841808,
+      "grad_norm": 2.3272323608398438,
+      "learning_rate": 4.5603349828701945e-06,
+      "loss": 1.5472,
+      "step": 539
+    },
+    {
+      "epoch": 0.5546995377503852,
+      "grad_norm": 2.416072130203247,
+      "learning_rate": 4.558431671107728e-06,
+      "loss": 1.5842,
+      "step": 540
+    },
+    {
+      "epoch": 0.5546995377503852,
+      "eval_loss": 1.825427532196045,
+      "eval_runtime": 37.9905,
+      "eval_samples_per_second": 26.322,
+      "eval_steps_per_second": 1.106,
+      "step": 540
+    },
+    {
+      "epoch": 0.5557267591165896,
+      "grad_norm": 2.3225271701812744,
+      "learning_rate": 4.556528359345261e-06,
+      "loss": 1.5188,
+      "step": 541
+    },
+    {
+      "epoch": 0.556753980482794,
+      "grad_norm": 2.2387030124664307,
+      "learning_rate": 4.5546250475827946e-06,
+      "loss": 1.4778,
+      "step": 542
+    },
+    {
+      "epoch": 0.5577812018489985,
+      "grad_norm": 2.3054494857788086,
+      "learning_rate": 4.552721735820328e-06,
+      "loss": 1.5229,
+      "step": 543
+    },
+    {
+      "epoch": 0.5588084232152029,
+      "grad_norm": 2.4956626892089844,
+      "learning_rate": 4.550818424057861e-06,
+      "loss": 1.5367,
+      "step": 544
+    },
+    {
+      "epoch": 0.5598356445814073,
+      "grad_norm": 2.454986810684204,
+      "learning_rate": 4.548915112295395e-06,
+      "loss": 1.5444,
+      "step": 545
+    },
+    {
+      "epoch": 0.5608628659476117,
+      "grad_norm": 2.466810703277588,
+      "learning_rate": 4.547011800532928e-06,
+      "loss": 1.4914,
+      "step": 546
+    },
+    {
+      "epoch": 0.5618900873138162,
+      "grad_norm": 2.51777982711792,
+      "learning_rate": 4.545108488770461e-06,
+      "loss": 1.5775,
+      "step": 547
+    },
+    {
+      "epoch": 0.5629173086800205,
+      "grad_norm": 2.340421199798584,
+      "learning_rate": 4.543205177007995e-06,
+      "loss": 1.5185,
+      "step": 548
+    },
+    {
+      "epoch": 0.5639445300462249,
+      "grad_norm": 2.376772880554199,
+      "learning_rate": 4.541301865245527e-06,
+      "loss": 1.5588,
+      "step": 549
+    },
+    {
+      "epoch": 0.5649717514124294,
+      "grad_norm": 2.4004335403442383,
+      "learning_rate": 4.539398553483061e-06,
+      "loss": 1.5283,
+      "step": 550
+    },
+    {
+      "epoch": 0.5649717514124294,
+      "eval_loss": 1.8294963836669922,
+      "eval_runtime": 37.9391,
+      "eval_samples_per_second": 26.358,
+      "eval_steps_per_second": 1.107,
+      "step": 550
+    },
+    {
+      "epoch": 0.5659989727786338,
+      "grad_norm": 2.424459934234619,
+      "learning_rate": 4.537495241720594e-06,
+      "loss": 1.5576,
+      "step": 551
+    },
+    {
+      "epoch": 0.5670261941448382,
+      "grad_norm": 2.4413998126983643,
+      "learning_rate": 4.535591929958127e-06,
+      "loss": 1.5444,
+      "step": 552
+    },
+    {
+      "epoch": 0.5680534155110426,
+      "grad_norm": 2.408897876739502,
+      "learning_rate": 4.533688618195661e-06,
+      "loss": 1.5269,
+      "step": 553
+    },
+    {
+      "epoch": 0.5690806368772471,
+      "grad_norm": 2.471024990081787,
+      "learning_rate": 4.531785306433194e-06,
+      "loss": 1.5109,
+      "step": 554
+    },
+    {
+      "epoch": 0.5701078582434514,
+      "grad_norm": 2.374882459640503,
+      "learning_rate": 4.529881994670727e-06,
+      "loss": 1.5205,
+      "step": 555
+    },
+    {
+      "epoch": 0.5711350796096559,
+      "grad_norm": 2.4166030883789062,
+      "learning_rate": 4.527978682908261e-06,
+      "loss": 1.5276,
+      "step": 556
+    },
+    {
+      "epoch": 0.5721623009758603,
+      "grad_norm": 2.2681620121002197,
+      "learning_rate": 4.526075371145794e-06,
+      "loss": 1.5374,
+      "step": 557
+    },
+    {
+      "epoch": 0.5731895223420647,
+      "grad_norm": 2.3933286666870117,
+      "learning_rate": 4.5241720593833275e-06,
+      "loss": 1.5349,
+      "step": 558
+    },
+    {
+      "epoch": 0.5742167437082691,
+      "grad_norm": 2.4086451530456543,
+      "learning_rate": 4.52226874762086e-06,
+      "loss": 1.5512,
+      "step": 559
+    },
+    {
+      "epoch": 0.5752439650744735,
+      "grad_norm": 2.559021472930908,
+      "learning_rate": 4.520365435858393e-06,
+      "loss": 1.5789,
+      "step": 560
+    },
+    {
+      "epoch": 0.5752439650744735,
+      "eval_loss": 1.8243234157562256,
+      "eval_runtime": 37.934,
+      "eval_samples_per_second": 26.362,
+      "eval_steps_per_second": 1.107,
+      "step": 560
+    },
+    {
+      "epoch": 0.576271186440678,
+      "grad_norm": 2.483494520187378,
+      "learning_rate": 4.518462124095928e-06,
+      "loss": 1.5348,
+      "step": 561
+    },
+    {
+      "epoch": 0.5772984078068824,
+      "grad_norm": 2.4039993286132812,
+      "learning_rate": 4.516558812333461e-06,
+      "loss": 1.5201,
+      "step": 562
+    },
+    {
+      "epoch": 0.5783256291730868,
+      "grad_norm": 2.5080959796905518,
+      "learning_rate": 4.514655500570994e-06,
+      "loss": 1.5864,
+      "step": 563
+    },
+    {
+      "epoch": 0.5793528505392912,
+      "grad_norm": 2.620887279510498,
+      "learning_rate": 4.512752188808528e-06,
+      "loss": 1.5901,
+      "step": 564
+    },
+    {
+      "epoch": 0.5803800719054957,
+      "grad_norm": 2.5691096782684326,
+      "learning_rate": 4.510848877046061e-06,
+      "loss": 1.5506,
+      "step": 565
+    },
+    {
+      "epoch": 0.5814072932717,
+      "grad_norm": 2.3777835369110107,
+      "learning_rate": 4.5089455652835936e-06,
+      "loss": 1.5149,
+      "step": 566
+    },
+    {
+      "epoch": 0.5824345146379045,
+      "grad_norm": 2.2791666984558105,
+      "learning_rate": 4.507042253521127e-06,
+      "loss": 1.4771,
+      "step": 567
+    },
+    {
+      "epoch": 0.5834617360041089,
+      "grad_norm": 2.4953627586364746,
+      "learning_rate": 4.50513894175866e-06,
+      "loss": 1.556,
+      "step": 568
+    },
+    {
+      "epoch": 0.5844889573703133,
+      "grad_norm": 2.3563215732574463,
+      "learning_rate": 4.503235629996194e-06,
+      "loss": 1.549,
+      "step": 569
+    },
+    {
+      "epoch": 0.5855161787365177,
+      "grad_norm": 2.49676513671875,
+      "learning_rate": 4.501332318233727e-06,
+      "loss": 1.513,
+      "step": 570
+    },
+    {
+      "epoch": 0.5855161787365177,
+      "eval_loss": 1.8353043794631958,
+      "eval_runtime": 37.9464,
+      "eval_samples_per_second": 26.353,
+      "eval_steps_per_second": 1.107,
+      "step": 570
+    },
+    {
+      "epoch": 0.5865434001027221,
+      "grad_norm": 2.385275363922119,
+      "learning_rate": 4.49942900647126e-06,
+      "loss": 1.5622,
+      "step": 571
+    },
+    {
+      "epoch": 0.5875706214689266,
+      "grad_norm": 2.3089914321899414,
+      "learning_rate": 4.497525694708794e-06,
+      "loss": 1.5426,
+      "step": 572
+    },
+    {
+      "epoch": 0.588597842835131,
+      "grad_norm": 2.307323455810547,
+      "learning_rate": 4.495622382946327e-06,
+      "loss": 1.5088,
+      "step": 573
+    },
+    {
+      "epoch": 0.5896250642013354,
+      "grad_norm": 2.597288131713867,
+      "learning_rate": 4.4937190711838605e-06,
+      "loss": 1.5788,
+      "step": 574
+    },
+    {
+      "epoch": 0.5906522855675398,
+      "grad_norm": 2.3923256397247314,
+      "learning_rate": 4.491815759421394e-06,
+      "loss": 1.5038,
+      "step": 575
+    },
+    {
+      "epoch": 0.5916795069337443,
+      "grad_norm": 2.4199156761169434,
+      "learning_rate": 4.489912447658926e-06,
+      "loss": 1.5057,
+      "step": 576
+    },
+    {
+      "epoch": 0.5927067282999486,
+      "grad_norm": 2.5643393993377686,
+      "learning_rate": 4.48800913589646e-06,
+      "loss": 1.5274,
+      "step": 577
+    },
+    {
+      "epoch": 0.593733949666153,
+      "grad_norm": 2.5854485034942627,
+      "learning_rate": 4.486105824133993e-06,
+      "loss": 1.5341,
+      "step": 578
+    },
+    {
+      "epoch": 0.5947611710323575,
+      "grad_norm": 2.5363597869873047,
+      "learning_rate": 4.4842025123715264e-06,
+      "loss": 1.4953,
+      "step": 579
+    },
+    {
+      "epoch": 0.5957883923985619,
+      "grad_norm": 2.3880693912506104,
+      "learning_rate": 4.48229920060906e-06,
+      "loss": 1.4868,
+      "step": 580
+    },
+    {
+      "epoch": 0.5957883923985619,
+      "eval_loss": 1.8251410722732544,
+      "eval_runtime": 37.9554,
+      "eval_samples_per_second": 26.347,
+      "eval_steps_per_second": 1.107,
+      "step": 580
+    },
+    {
+      "epoch": 0.5968156137647663,
+      "grad_norm": 2.5272583961486816,
+      "learning_rate": 4.480395888846594e-06,
+      "loss": 1.5851,
+      "step": 581
+    },
+    {
+      "epoch": 0.5978428351309707,
+      "grad_norm": 2.4496943950653076,
+      "learning_rate": 4.478492577084127e-06,
+      "loss": 1.4896,
+      "step": 582
+    },
+    {
+      "epoch": 0.5988700564971752,
+      "grad_norm": 2.3996055126190186,
+      "learning_rate": 4.47658926532166e-06,
+      "loss": 1.5253,
+      "step": 583
+    },
+    {
+      "epoch": 0.5998972778633795,
+      "grad_norm": 2.5964202880859375,
+      "learning_rate": 4.474685953559193e-06,
+      "loss": 1.6294,
+      "step": 584
+    },
+    {
+      "epoch": 0.600924499229584,
+      "grad_norm": 2.5567739009857178,
+      "learning_rate": 4.472782641796727e-06,
+      "loss": 1.4834,
+      "step": 585
+    },
+    {
+      "epoch": 0.6019517205957884,
+      "grad_norm": 2.440291166305542,
+      "learning_rate": 4.47087933003426e-06,
+      "loss": 1.4966,
+      "step": 586
+    },
+    {
+      "epoch": 0.6029789419619929,
+      "grad_norm": 2.525763988494873,
+      "learning_rate": 4.468976018271793e-06,
+      "loss": 1.5005,
+      "step": 587
+    },
+    {
+      "epoch": 0.6040061633281972,
+      "grad_norm": 2.4822962284088135,
+      "learning_rate": 4.467072706509327e-06,
+      "loss": 1.5358,
+      "step": 588
+    },
+    {
+      "epoch": 0.6050333846944016,
+      "grad_norm": 2.591149091720581,
+      "learning_rate": 4.46516939474686e-06,
+      "loss": 1.4975,
+      "step": 589
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "grad_norm": 2.4884238243103027,
+      "learning_rate": 4.463266082984393e-06,
+      "loss": 1.6107,
+      "step": 590
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "eval_loss": 1.8383466005325317,
+      "eval_runtime": 37.9554,
+      "eval_samples_per_second": 26.347,
+      "eval_steps_per_second": 1.107,
+      "step": 590
+    },
+    {
+      "epoch": 0.6070878274268104,
+      "grad_norm": 2.4965732097625732,
+      "learning_rate": 4.461362771221927e-06,
+      "loss": 1.4986,
+      "step": 591
+    },
+    {
+      "epoch": 0.6081150487930149,
+      "grad_norm": 2.544792413711548,
+      "learning_rate": 4.45945945945946e-06,
+      "loss": 1.5067,
+      "step": 592
+    },
+    {
+      "epoch": 0.6091422701592193,
+      "grad_norm": 2.5726170539855957,
+      "learning_rate": 4.457556147696993e-06,
+      "loss": 1.5861,
+      "step": 593
+    },
+    {
+      "epoch": 0.6101694915254238,
+      "grad_norm": 2.493027448654175,
+      "learning_rate": 4.455652835934526e-06,
+      "loss": 1.5857,
+      "step": 594
+    },
+    {
+      "epoch": 0.6111967128916281,
+      "grad_norm": 2.470771074295044,
+      "learning_rate": 4.453749524172059e-06,
+      "loss": 1.5419,
+      "step": 595
+    },
+    {
+      "epoch": 0.6122239342578326,
+      "grad_norm": 2.568052053451538,
+      "learning_rate": 4.451846212409593e-06,
+      "loss": 1.5391,
+      "step": 596
+    },
+    {
+      "epoch": 0.613251155624037,
+      "grad_norm": 2.4346213340759277,
+      "learning_rate": 4.449942900647126e-06,
+      "loss": 1.4761,
+      "step": 597
+    },
+    {
+      "epoch": 0.6142783769902413,
+      "grad_norm": 2.418133497238159,
+      "learning_rate": 4.4480395888846595e-06,
+      "loss": 1.5383,
+      "step": 598
+    },
+    {
+      "epoch": 0.6153055983564458,
+      "grad_norm": 2.4795567989349365,
+      "learning_rate": 4.446136277122193e-06,
+      "loss": 1.4777,
+      "step": 599
+    },
+    {
+      "epoch": 0.6163328197226502,
+      "grad_norm": 2.5122742652893066,
+      "learning_rate": 4.444232965359726e-06,
+      "loss": 1.5353,
+      "step": 600
+    },
+    {
+      "epoch": 0.6163328197226502,
+      "eval_loss": 1.8325241804122925,
+      "eval_runtime": 37.9546,
+      "eval_samples_per_second": 26.347,
+      "eval_steps_per_second": 1.107,
+      "step": 600
+    },
+    {
+      "epoch": 0.6173600410888547,
+      "grad_norm": 2.604067325592041,
+      "learning_rate": 4.4423296535972596e-06,
+      "loss": 1.489,
+      "step": 601
+    },
+    {
+      "epoch": 0.618387262455059,
+      "grad_norm": 2.3741753101348877,
+      "learning_rate": 4.440426341834793e-06,
+      "loss": 1.532,
+      "step": 602
+    },
+    {
+      "epoch": 0.6194144838212635,
+      "grad_norm": 2.5373036861419678,
+      "learning_rate": 4.438523030072326e-06,
+      "loss": 1.561,
+      "step": 603
+    },
+    {
+      "epoch": 0.6204417051874679,
+      "grad_norm": 2.2958099842071533,
+      "learning_rate": 4.43661971830986e-06,
+      "loss": 1.5025,
+      "step": 604
+    },
+    {
+      "epoch": 0.6214689265536724,
+      "grad_norm": 2.417006015777588,
+      "learning_rate": 4.434716406547393e-06,
+      "loss": 1.4977,
+      "step": 605
+    },
+    {
+      "epoch": 0.6224961479198767,
+      "grad_norm": 2.335296392440796,
+      "learning_rate": 4.432813094784926e-06,
+      "loss": 1.5504,
+      "step": 606
+    },
+    {
+      "epoch": 0.6235233692860811,
+      "grad_norm": 2.4086828231811523,
+      "learning_rate": 4.43090978302246e-06,
+      "loss": 1.5217,
+      "step": 607
+    },
+    {
+      "epoch": 0.6245505906522856,
+      "grad_norm": 2.3816819190979004,
+      "learning_rate": 4.429006471259993e-06,
+      "loss": 1.5126,
+      "step": 608
+    },
+    {
+      "epoch": 0.6255778120184899,
+      "grad_norm": 2.307671546936035,
+      "learning_rate": 4.4271031594975265e-06,
+      "loss": 1.4876,
+      "step": 609
+    },
+    {
+      "epoch": 0.6266050333846944,
+      "grad_norm": 2.355837821960449,
+      "learning_rate": 4.425199847735059e-06,
+      "loss": 1.4827,
+      "step": 610
+    },
+    {
+      "epoch": 0.6266050333846944,
+      "eval_loss": 1.8269455432891846,
+      "eval_runtime": 37.6459,
+      "eval_samples_per_second": 26.563,
+      "eval_steps_per_second": 1.116,
+      "step": 610
+    },
+    {
+      "epoch": 0.6276322547508988,
+      "grad_norm": 2.558631658554077,
+      "learning_rate": 4.423296535972592e-06,
+      "loss": 1.5172,
+      "step": 611
+    },
+    {
+      "epoch": 0.6286594761171033,
+      "grad_norm": 2.3170175552368164,
+      "learning_rate": 4.421393224210126e-06,
+      "loss": 1.4784,
+      "step": 612
+    },
+    {
+      "epoch": 0.6296866974833076,
+      "grad_norm": 2.4064524173736572,
+      "learning_rate": 4.419489912447659e-06,
+      "loss": 1.5616,
+      "step": 613
+    },
+    {
+      "epoch": 0.6307139188495121,
+      "grad_norm": 2.3719160556793213,
+      "learning_rate": 4.4175866006851924e-06,
+      "loss": 1.5292,
+      "step": 614
+    },
+    {
+      "epoch": 0.6317411402157165,
+      "grad_norm": 2.3808202743530273,
+      "learning_rate": 4.415683288922726e-06,
+      "loss": 1.5122,
+      "step": 615
+    },
+    {
+      "epoch": 0.632768361581921,
+      "grad_norm": 2.4208037853240967,
+      "learning_rate": 4.413779977160259e-06,
+      "loss": 1.5088,
+      "step": 616
+    },
+    {
+      "epoch": 0.6337955829481253,
+      "grad_norm": 2.5558979511260986,
+      "learning_rate": 4.4118766653977925e-06,
+      "loss": 1.5678,
+      "step": 617
+    },
+    {
+      "epoch": 0.6348228043143297,
+      "grad_norm": 2.443908452987671,
+      "learning_rate": 4.409973353635326e-06,
+      "loss": 1.5234,
+      "step": 618
+    },
+    {
+      "epoch": 0.6358500256805342,
+      "grad_norm": 2.420408010482788,
+      "learning_rate": 4.408070041872859e-06,
+      "loss": 1.4708,
+      "step": 619
+    },
+    {
+      "epoch": 0.6368772470467385,
+      "grad_norm": 2.3884904384613037,
+      "learning_rate": 4.406166730110393e-06,
+      "loss": 1.4821,
+      "step": 620
+    },
+    {
+      "epoch": 0.6368772470467385,
+      "eval_loss": 1.8217289447784424,
+      "eval_runtime": 37.7936,
+      "eval_samples_per_second": 26.459,
+      "eval_steps_per_second": 1.111,
+      "step": 620
+    },
+    {
+      "epoch": 0.637904468412943,
+      "grad_norm": 2.313720703125,
+      "learning_rate": 4.404263418347925e-06,
+      "loss": 1.4962,
+      "step": 621
+    },
+    {
+      "epoch": 0.6389316897791474,
+      "grad_norm": 2.3173866271972656,
+      "learning_rate": 4.402360106585459e-06,
+      "loss": 1.4857,
+      "step": 622
+    },
+    {
+      "epoch": 0.6399589111453519,
+      "grad_norm": 2.541757822036743,
+      "learning_rate": 4.400456794822993e-06,
+      "loss": 1.5177,
+      "step": 623
+    },
+    {
+      "epoch": 0.6409861325115562,
+      "grad_norm": 2.4382357597351074,
+      "learning_rate": 4.398553483060526e-06,
+      "loss": 1.4718,
+      "step": 624
+    },
+    {
+      "epoch": 0.6420133538777607,
+      "grad_norm": 2.2988059520721436,
+      "learning_rate": 4.396650171298059e-06,
+      "loss": 1.4883,
+      "step": 625
+    },
+    {
+      "epoch": 0.6430405752439651,
+      "grad_norm": 2.4817495346069336,
+      "learning_rate": 4.394746859535593e-06,
+      "loss": 1.4823,
+      "step": 626
+    },
+    {
+      "epoch": 0.6440677966101694,
+      "grad_norm": 2.4578585624694824,
+      "learning_rate": 4.392843547773125e-06,
+      "loss": 1.4882,
+      "step": 627
+    },
+    {
+      "epoch": 0.6450950179763739,
+      "grad_norm": 2.184972047805786,
+      "learning_rate": 4.390940236010659e-06,
+      "loss": 1.514,
+      "step": 628
+    },
+    {
+      "epoch": 0.6461222393425783,
+      "grad_norm": 2.4040279388427734,
+      "learning_rate": 4.389036924248192e-06,
+      "loss": 1.5343,
+      "step": 629
+    },
+    {
+      "epoch": 0.6471494607087828,
+      "grad_norm": 2.3676459789276123,
+      "learning_rate": 4.387133612485725e-06,
+      "loss": 1.5007,
+      "step": 630
+    },
+    {
+      "epoch": 0.6471494607087828,
+      "eval_loss": 1.8106164932250977,
+      "eval_runtime": 37.9383,
+      "eval_samples_per_second": 26.359,
+      "eval_steps_per_second": 1.107,
+      "step": 630
+    },
+    {
+      "epoch": 0.6481766820749871,
+      "grad_norm": 2.234689235687256,
+      "learning_rate": 4.385230300723259e-06,
+      "loss": 1.5064,
+      "step": 631
+    },
+    {
+      "epoch": 0.6492039034411916,
+      "grad_norm": 2.281742572784424,
+      "learning_rate": 4.383326988960792e-06,
+      "loss": 1.4777,
+      "step": 632
+    },
+    {
+      "epoch": 0.650231124807396,
+      "grad_norm": 2.4314322471618652,
+      "learning_rate": 4.3814236771983255e-06,
+      "loss": 1.5331,
+      "step": 633
+    },
+    {
+      "epoch": 0.6512583461736005,
+      "grad_norm": 2.4092588424682617,
+      "learning_rate": 4.379520365435859e-06,
+      "loss": 1.5122,
+      "step": 634
+    },
+    {
+      "epoch": 0.6522855675398048,
+      "grad_norm": 2.517362117767334,
+      "learning_rate": 4.377617053673392e-06,
+      "loss": 1.52,
+      "step": 635
+    },
+    {
+      "epoch": 0.6533127889060092,
+      "grad_norm": 2.4784109592437744,
+      "learning_rate": 4.3757137419109256e-06,
+      "loss": 1.5771,
+      "step": 636
+    },
+    {
+      "epoch": 0.6543400102722137,
+      "grad_norm": 2.32181715965271,
+      "learning_rate": 4.373810430148459e-06,
+      "loss": 1.4718,
+      "step": 637
+    },
+    {
+      "epoch": 0.655367231638418,
+      "grad_norm": 2.5146260261535645,
+      "learning_rate": 4.3719071183859914e-06,
+      "loss": 1.5307,
+      "step": 638
+    },
+    {
+      "epoch": 0.6563944530046225,
+      "grad_norm": 2.2364494800567627,
+      "learning_rate": 4.370003806623525e-06,
+      "loss": 1.4803,
+      "step": 639
+    },
+    {
+      "epoch": 0.6574216743708269,
+      "grad_norm": 2.3735806941986084,
+      "learning_rate": 4.368100494861058e-06,
+      "loss": 1.5249,
+      "step": 640
+    },
+    {
+      "epoch": 0.6574216743708269,
+      "eval_loss": 1.8134880065917969,
+      "eval_runtime": 38.0376,
+      "eval_samples_per_second": 26.29,
+      "eval_steps_per_second": 1.104,
+      "step": 640
+    },
+    {
+      "epoch": 0.6584488957370314,
+      "grad_norm": 2.4600815773010254,
+      "learning_rate": 4.3661971830985915e-06,
+      "loss": 1.4893,
+      "step": 641
+    },
+    {
+      "epoch": 0.6594761171032357,
+      "grad_norm": 2.4106783866882324,
+      "learning_rate": 4.364293871336126e-06,
+      "loss": 1.5163,
+      "step": 642
+    },
+    {
+      "epoch": 0.6605033384694402,
+      "grad_norm": 2.276624917984009,
+      "learning_rate": 4.362390559573659e-06,
+      "loss": 1.482,
+      "step": 643
+    },
+    {
+      "epoch": 0.6615305598356446,
+      "grad_norm": 2.336341381072998,
+      "learning_rate": 4.3604872478111925e-06,
+      "loss": 1.4152,
+      "step": 644
+    },
+    {
+      "epoch": 0.662557781201849,
+      "grad_norm": 2.3329105377197266,
+      "learning_rate": 4.358583936048725e-06,
+      "loss": 1.5411,
+      "step": 645
+    },
+    {
+      "epoch": 0.6635850025680534,
+      "grad_norm": 2.2877161502838135,
+      "learning_rate": 4.356680624286258e-06,
+      "loss": 1.4997,
+      "step": 646
+    },
+    {
+      "epoch": 0.6646122239342578,
+      "grad_norm": 2.410820484161377,
+      "learning_rate": 4.354777312523792e-06,
+      "loss": 1.5446,
+      "step": 647
+    },
+    {
+      "epoch": 0.6656394453004623,
+      "grad_norm": 2.403038501739502,
+      "learning_rate": 4.352874000761325e-06,
+      "loss": 1.4916,
+      "step": 648
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 2.3261778354644775,
+      "learning_rate": 4.3509706889988584e-06,
+      "loss": 1.5561,
+      "step": 649
+    },
+    {
+      "epoch": 0.6676938880328711,
+      "grad_norm": 2.4512174129486084,
+      "learning_rate": 4.349067377236392e-06,
+      "loss": 1.556,
+      "step": 650
+    },
+    {
+      "epoch": 0.6676938880328711,
+      "eval_loss": 1.8238279819488525,
+      "eval_runtime": 37.9988,
+      "eval_samples_per_second": 26.317,
+      "eval_steps_per_second": 1.105,
+      "step": 650
+    },
+    {
+      "epoch": 0.6687211093990755,
+      "grad_norm": 2.4523823261260986,
+      "learning_rate": 4.347164065473925e-06,
+      "loss": 1.5676,
+      "step": 651
+    },
+    {
+      "epoch": 0.66974833076528,
+      "grad_norm": 2.3770365715026855,
+      "learning_rate": 4.3452607537114585e-06,
+      "loss": 1.4815,
+      "step": 652
+    },
+    {
+      "epoch": 0.6707755521314843,
+      "grad_norm": 2.2440760135650635,
+      "learning_rate": 4.343357441948992e-06,
+      "loss": 1.5141,
+      "step": 653
+    },
+    {
+      "epoch": 0.6718027734976888,
+      "grad_norm": 2.3341948986053467,
+      "learning_rate": 4.341454130186525e-06,
+      "loss": 1.4859,
+      "step": 654
+    },
+    {
+      "epoch": 0.6728299948638932,
+      "grad_norm": 2.507354497909546,
+      "learning_rate": 4.339550818424058e-06,
+      "loss": 1.4734,
+      "step": 655
+    },
+    {
+      "epoch": 0.6738572162300975,
+      "grad_norm": 2.395954132080078,
+      "learning_rate": 4.337647506661591e-06,
+      "loss": 1.5069,
+      "step": 656
+    },
+    {
+      "epoch": 0.674884437596302,
+      "grad_norm": 2.362211227416992,
+      "learning_rate": 4.3357441948991245e-06,
+      "loss": 1.5299,
+      "step": 657
+    },
+    {
+      "epoch": 0.6759116589625064,
+      "grad_norm": 2.460008144378662,
+      "learning_rate": 4.333840883136658e-06,
+      "loss": 1.4795,
+      "step": 658
+    },
+    {
+      "epoch": 0.6769388803287109,
+      "grad_norm": 2.3578217029571533,
+      "learning_rate": 4.331937571374191e-06,
+      "loss": 1.4906,
+      "step": 659
+    },
+    {
+      "epoch": 0.6779661016949152,
+      "grad_norm": 2.398843288421631,
+      "learning_rate": 4.3300342596117246e-06,
+      "loss": 1.4986,
+      "step": 660
+    },
+    {
+      "epoch": 0.6779661016949152,
+      "eval_loss": 1.8206881284713745,
+      "eval_runtime": 38.044,
+      "eval_samples_per_second": 26.285,
+      "eval_steps_per_second": 1.104,
+      "step": 660
+    },
+    {
+      "epoch": 0.6789933230611197,
+      "grad_norm": 2.3587260246276855,
+      "learning_rate": 4.328130947849258e-06,
+      "loss": 1.461,
+      "step": 661
+    },
+    {
+      "epoch": 0.6800205444273241,
+      "grad_norm": 2.311952590942383,
+      "learning_rate": 4.326227636086791e-06,
+      "loss": 1.5041,
+      "step": 662
+    },
+    {
+      "epoch": 0.6810477657935285,
+      "grad_norm": 2.376188278198242,
+      "learning_rate": 4.324324324324325e-06,
+      "loss": 1.5042,
+      "step": 663
+    },
+    {
+      "epoch": 0.6820749871597329,
+      "grad_norm": 2.5807230472564697,
+      "learning_rate": 4.322421012561858e-06,
+      "loss": 1.4935,
+      "step": 664
+    },
+    {
+      "epoch": 0.6831022085259374,
+      "grad_norm": 2.368128538131714,
+      "learning_rate": 4.320517700799391e-06,
+      "loss": 1.5851,
+      "step": 665
+    },
+    {
+      "epoch": 0.6841294298921418,
+      "grad_norm": 2.4517524242401123,
+      "learning_rate": 4.318614389036925e-06,
+      "loss": 1.5528,
+      "step": 666
+    },
+    {
+      "epoch": 0.6851566512583461,
+      "grad_norm": 2.5436811447143555,
+      "learning_rate": 4.316711077274458e-06,
+      "loss": 1.4633,
+      "step": 667
+    },
+    {
+      "epoch": 0.6861838726245506,
+      "grad_norm": 2.4287915229797363,
+      "learning_rate": 4.3148077655119915e-06,
+      "loss": 1.5454,
+      "step": 668
+    },
+    {
+      "epoch": 0.687211093990755,
+      "grad_norm": 2.386528730392456,
+      "learning_rate": 4.312904453749525e-06,
+      "loss": 1.5116,
+      "step": 669
+    },
+    {
+      "epoch": 0.6882383153569595,
+      "grad_norm": 2.6602251529693604,
+      "learning_rate": 4.311001141987058e-06,
+      "loss": 1.5199,
+      "step": 670
+    },
+    {
+      "epoch": 0.6882383153569595,
+      "eval_loss": 1.8252663612365723,
+      "eval_runtime": 38.0791,
+      "eval_samples_per_second": 26.261,
+      "eval_steps_per_second": 1.103,
+      "step": 670
+    },
+    {
+      "epoch": 0.6892655367231638,
+      "grad_norm": 2.351128339767456,
+      "learning_rate": 4.3090978302245916e-06,
+      "loss": 1.507,
+      "step": 671
+    },
+    {
+      "epoch": 0.6902927580893683,
+      "grad_norm": 2.393441677093506,
+      "learning_rate": 4.307194518462124e-06,
+      "loss": 1.579,
+      "step": 672
+    },
+    {
+      "epoch": 0.6913199794555727,
+      "grad_norm": 2.3730578422546387,
+      "learning_rate": 4.3052912066996574e-06,
+      "loss": 1.5446,
+      "step": 673
+    },
+    {
+      "epoch": 0.6923472008217771,
+      "grad_norm": 2.2522945404052734,
+      "learning_rate": 4.303387894937191e-06,
+      "loss": 1.4705,
+      "step": 674
+    },
+    {
+      "epoch": 0.6933744221879815,
+      "grad_norm": 2.384563446044922,
+      "learning_rate": 4.301484583174724e-06,
+      "loss": 1.4636,
+      "step": 675
+    },
+    {
+      "epoch": 0.6944016435541859,
+      "grad_norm": 2.2879116535186768,
+      "learning_rate": 4.2995812714122575e-06,
+      "loss": 1.4914,
+      "step": 676
+    },
+    {
+      "epoch": 0.6954288649203904,
+      "grad_norm": 2.494788646697998,
+      "learning_rate": 4.297677959649791e-06,
+      "loss": 1.5126,
+      "step": 677
+    },
+    {
+      "epoch": 0.6964560862865947,
+      "grad_norm": 2.700012683868408,
+      "learning_rate": 4.295774647887324e-06,
+      "loss": 1.5362,
+      "step": 678
+    },
+    {
+      "epoch": 0.6974833076527992,
+      "grad_norm": 2.5130302906036377,
+      "learning_rate": 4.293871336124858e-06,
+      "loss": 1.5075,
+      "step": 679
+    },
+    {
+      "epoch": 0.6985105290190036,
+      "grad_norm": 2.486236572265625,
+      "learning_rate": 4.291968024362391e-06,
+      "loss": 1.5338,
+      "step": 680
+    },
+    {
+      "epoch": 0.6985105290190036,
+      "eval_loss": 1.8229657411575317,
+      "eval_runtime": 38.0642,
+      "eval_samples_per_second": 26.271,
+      "eval_steps_per_second": 1.103,
+      "step": 680
+    },
+    {
+      "epoch": 0.699537750385208,
+      "grad_norm": 2.441683530807495,
+      "learning_rate": 4.290064712599924e-06,
+      "loss": 1.537,
+      "step": 681
+    },
+    {
+      "epoch": 0.7005649717514124,
+      "grad_norm": 2.396798610687256,
+      "learning_rate": 4.288161400837458e-06,
+      "loss": 1.4553,
+      "step": 682
+    },
+    {
+      "epoch": 0.7015921931176169,
+      "grad_norm": 2.3636221885681152,
+      "learning_rate": 4.286258089074991e-06,
+      "loss": 1.5138,
+      "step": 683
+    },
+    {
+      "epoch": 0.7026194144838213,
+      "grad_norm": 2.4584708213806152,
+      "learning_rate": 4.284354777312524e-06,
+      "loss": 1.4725,
+      "step": 684
+    },
+    {
+      "epoch": 0.7036466358500256,
+      "grad_norm": 2.850332498550415,
+      "learning_rate": 4.282451465550058e-06,
+      "loss": 1.5075,
+      "step": 685
+    },
+    {
+      "epoch": 0.7046738572162301,
+      "grad_norm": 2.6345887184143066,
+      "learning_rate": 4.280548153787591e-06,
+      "loss": 1.5054,
+      "step": 686
+    },
+    {
+      "epoch": 0.7057010785824345,
+      "grad_norm": 2.3535311222076416,
+      "learning_rate": 4.2786448420251245e-06,
+      "loss": 1.4607,
+      "step": 687
+    },
+    {
+      "epoch": 0.706728299948639,
+      "grad_norm": 2.490431547164917,
+      "learning_rate": 4.276741530262658e-06,
+      "loss": 1.4852,
+      "step": 688
+    },
+    {
+      "epoch": 0.7077555213148433,
+      "grad_norm": 2.4353036880493164,
+      "learning_rate": 4.27483821850019e-06,
+      "loss": 1.5503,
+      "step": 689
+    },
+    {
+      "epoch": 0.7087827426810478,
+      "grad_norm": 2.3356940746307373,
+      "learning_rate": 4.272934906737724e-06,
+      "loss": 1.5416,
+      "step": 690
+    },
+    {
+      "epoch": 0.7087827426810478,
+      "eval_loss": 1.8164218664169312,
+      "eval_runtime": 38.0592,
+      "eval_samples_per_second": 26.275,
+      "eval_steps_per_second": 1.104,
+      "step": 690
+    },
+    {
+      "epoch": 0.7098099640472522,
+      "grad_norm": 2.4772775173187256,
+      "learning_rate": 4.271031594975257e-06,
+      "loss": 1.5392,
+      "step": 691
+    },
+    {
+      "epoch": 0.7108371854134566,
+      "grad_norm": 2.2921438217163086,
+      "learning_rate": 4.2691282832127905e-06,
+      "loss": 1.4935,
+      "step": 692
+    },
+    {
+      "epoch": 0.711864406779661,
+      "grad_norm": 2.3176305294036865,
+      "learning_rate": 4.267224971450324e-06,
+      "loss": 1.4451,
+      "step": 693
+    },
+    {
+      "epoch": 0.7128916281458655,
+      "grad_norm": 2.6825685501098633,
+      "learning_rate": 4.265321659687857e-06,
+      "loss": 1.4878,
+      "step": 694
+    },
+    {
+      "epoch": 0.7139188495120699,
+      "grad_norm": 2.6141414642333984,
+      "learning_rate": 4.2634183479253906e-06,
+      "loss": 1.4695,
+      "step": 695
+    },
+    {
+      "epoch": 0.7149460708782742,
+      "grad_norm": 2.5179429054260254,
+      "learning_rate": 4.261515036162924e-06,
+      "loss": 1.492,
+      "step": 696
+    },
+    {
+      "epoch": 0.7159732922444787,
+      "grad_norm": 2.4744884967803955,
+      "learning_rate": 4.259611724400457e-06,
+      "loss": 1.4752,
+      "step": 697
+    },
+    {
+      "epoch": 0.7170005136106831,
+      "grad_norm": 2.4911048412323,
+      "learning_rate": 4.257708412637991e-06,
+      "loss": 1.423,
+      "step": 698
+    },
+    {
+      "epoch": 0.7180277349768875,
+      "grad_norm": 2.5116214752197266,
+      "learning_rate": 4.255805100875523e-06,
+      "loss": 1.4535,
+      "step": 699
+    },
+    {
+      "epoch": 0.7190549563430919,
+      "grad_norm": 2.476651430130005,
+      "learning_rate": 4.2539017891130565e-06,
+      "loss": 1.5349,
+      "step": 700
+    },
+    {
+      "epoch": 0.7190549563430919,
+      "eval_loss": 1.8069179058074951,
+      "eval_runtime": 38.0292,
+      "eval_samples_per_second": 26.296,
+      "eval_steps_per_second": 1.104,
+      "step": 700
+    },
+    {
+      "epoch": 0.7200821777092964,
+      "grad_norm": 2.5234575271606445,
+      "learning_rate": 4.25199847735059e-06,
+      "loss": 1.4596,
+      "step": 701
+    },
+    {
+      "epoch": 0.7211093990755008,
+      "grad_norm": 2.33290433883667,
+      "learning_rate": 4.250095165588123e-06,
+      "loss": 1.5147,
+      "step": 702
+    },
+    {
+      "epoch": 0.7221366204417052,
+      "grad_norm": 2.479883909225464,
+      "learning_rate": 4.2481918538256575e-06,
+      "loss": 1.47,
+      "step": 703
+    },
+    {
+      "epoch": 0.7231638418079096,
+      "grad_norm": 2.432253837585449,
+      "learning_rate": 4.246288542063191e-06,
+      "loss": 1.5152,
+      "step": 704
+    },
+    {
+      "epoch": 0.724191063174114,
+      "grad_norm": 2.358304738998413,
+      "learning_rate": 4.244385230300724e-06,
+      "loss": 1.4562,
+      "step": 705
+    },
+    {
+      "epoch": 0.7252182845403184,
+      "grad_norm": 2.480877161026001,
+      "learning_rate": 4.242481918538257e-06,
+      "loss": 1.4906,
+      "step": 706
+    },
+    {
+      "epoch": 0.7262455059065228,
+      "grad_norm": 2.4287192821502686,
+      "learning_rate": 4.24057860677579e-06,
+      "loss": 1.5294,
+      "step": 707
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.4782888889312744,
+      "learning_rate": 4.2386752950133234e-06,
+      "loss": 1.4767,
+      "step": 708
+    },
+    {
+      "epoch": 0.7282999486389317,
+      "grad_norm": 2.4948253631591797,
+      "learning_rate": 4.236771983250857e-06,
+      "loss": 1.4664,
+      "step": 709
+    },
+    {
+      "epoch": 0.7293271700051361,
+      "grad_norm": 2.4625205993652344,
+      "learning_rate": 4.23486867148839e-06,
+      "loss": 1.495,
+      "step": 710
+    },
+    {
+      "epoch": 0.7293271700051361,
+      "eval_loss": 1.8140079975128174,
+      "eval_runtime": 37.6964,
+      "eval_samples_per_second": 26.528,
+      "eval_steps_per_second": 1.114,
+      "step": 710
+    },
+    {
+      "epoch": 0.7303543913713405,
+      "grad_norm": 2.669027805328369,
+      "learning_rate": 4.2329653597259235e-06,
+      "loss": 1.5204,
+      "step": 711
+    },
+    {
+      "epoch": 0.731381612737545,
+      "grad_norm": 2.5024707317352295,
+      "learning_rate": 4.231062047963457e-06,
+      "loss": 1.5283,
+      "step": 712
+    },
+    {
+      "epoch": 0.7324088341037494,
+      "grad_norm": 2.498258590698242,
+      "learning_rate": 4.22915873620099e-06,
+      "loss": 1.5139,
+      "step": 713
+    },
+    {
+      "epoch": 0.7334360554699538,
+      "grad_norm": 2.4305667877197266,
+      "learning_rate": 4.227255424438524e-06,
+      "loss": 1.5337,
+      "step": 714
+    },
+    {
+      "epoch": 0.7344632768361582,
+      "grad_norm": 2.3153419494628906,
+      "learning_rate": 4.225352112676057e-06,
+      "loss": 1.4646,
+      "step": 715
+    },
+    {
+      "epoch": 0.7354904982023626,
+      "grad_norm": 2.2835583686828613,
+      "learning_rate": 4.2234488009135895e-06,
+      "loss": 1.4328,
+      "step": 716
+    },
+    {
+      "epoch": 0.736517719568567,
+      "grad_norm": 2.401418924331665,
+      "learning_rate": 4.221545489151123e-06,
+      "loss": 1.4685,
+      "step": 717
+    },
+    {
+      "epoch": 0.7375449409347714,
+      "grad_norm": 2.454864978790283,
+      "learning_rate": 4.219642177388656e-06,
+      "loss": 1.4925,
+      "step": 718
+    },
+    {
+      "epoch": 0.7385721623009759,
+      "grad_norm": 2.2332873344421387,
+      "learning_rate": 4.2177388656261896e-06,
+      "loss": 1.4481,
+      "step": 719
+    },
+    {
+      "epoch": 0.7395993836671803,
+      "grad_norm": 2.46628475189209,
+      "learning_rate": 4.215835553863723e-06,
+      "loss": 1.4173,
+      "step": 720
+    },
+    {
+      "epoch": 0.7395993836671803,
+      "eval_loss": 1.817592740058899,
+      "eval_runtime": 37.8086,
+      "eval_samples_per_second": 26.449,
+      "eval_steps_per_second": 1.111,
+      "step": 720
+    },
+    {
+      "epoch": 0.7406266050333847,
+      "grad_norm": 2.346245765686035,
+      "learning_rate": 4.213932242101256e-06,
+      "loss": 1.4688,
+      "step": 721
+    },
+    {
+      "epoch": 0.7416538263995891,
+      "grad_norm": 2.2708640098571777,
+      "learning_rate": 4.21202893033879e-06,
+      "loss": 1.4937,
+      "step": 722
+    },
+    {
+      "epoch": 0.7426810477657936,
+      "grad_norm": 2.3716516494750977,
+      "learning_rate": 4.210125618576323e-06,
+      "loss": 1.4869,
+      "step": 723
+    },
+    {
+      "epoch": 0.743708269131998,
+      "grad_norm": 2.443058490753174,
+      "learning_rate": 4.208222306813856e-06,
+      "loss": 1.542,
+      "step": 724
+    },
+    {
+      "epoch": 0.7447354904982023,
+      "grad_norm": 2.327270984649658,
+      "learning_rate": 4.20631899505139e-06,
+      "loss": 1.3933,
+      "step": 725
+    },
+    {
+      "epoch": 0.7457627118644068,
+      "grad_norm": 2.3734512329101562,
+      "learning_rate": 4.204415683288923e-06,
+      "loss": 1.5109,
+      "step": 726
+    },
+    {
+      "epoch": 0.7467899332306112,
+      "grad_norm": 2.3631174564361572,
+      "learning_rate": 4.2025123715264565e-06,
+      "loss": 1.474,
+      "step": 727
+    },
+    {
+      "epoch": 0.7478171545968156,
+      "grad_norm": 2.306068181991577,
+      "learning_rate": 4.20060905976399e-06,
+      "loss": 1.42,
+      "step": 728
+    },
+    {
+      "epoch": 0.74884437596302,
+      "grad_norm": 2.391889810562134,
+      "learning_rate": 4.198705748001523e-06,
+      "loss": 1.5178,
+      "step": 729
+    },
+    {
+      "epoch": 0.7498715973292245,
+      "grad_norm": 2.3151705265045166,
+      "learning_rate": 4.1968024362390566e-06,
+      "loss": 1.4808,
+      "step": 730
+    },
+    {
+      "epoch": 0.7498715973292245,
+      "eval_loss": 1.8157070875167847,
+      "eval_runtime": 37.8799,
+      "eval_samples_per_second": 26.399,
+      "eval_steps_per_second": 1.109,
+      "step": 730
+    },
+    {
+      "epoch": 0.7508988186954288,
+      "grad_norm": 2.336479902267456,
+      "learning_rate": 4.19489912447659e-06,
+      "loss": 1.5438,
+      "step": 731
+    },
+    {
+      "epoch": 0.7519260400616333,
+      "grad_norm": 2.338242292404175,
+      "learning_rate": 4.192995812714123e-06,
+      "loss": 1.4869,
+      "step": 732
+    },
+    {
+      "epoch": 0.7529532614278377,
+      "grad_norm": 2.268650770187378,
+      "learning_rate": 4.191092500951656e-06,
+      "loss": 1.4787,
+      "step": 733
+    },
+    {
+      "epoch": 0.7539804827940421,
+      "grad_norm": 2.3010005950927734,
+      "learning_rate": 4.189189189189189e-06,
+      "loss": 1.4711,
+      "step": 734
+    },
+    {
+      "epoch": 0.7550077041602465,
+      "grad_norm": 2.54770827293396,
+      "learning_rate": 4.1872858774267225e-06,
+      "loss": 1.5351,
+      "step": 735
+    },
+    {
+      "epoch": 0.7560349255264509,
+      "grad_norm": 2.3412787914276123,
+      "learning_rate": 4.185382565664256e-06,
+      "loss": 1.4655,
+      "step": 736
+    },
+    {
+      "epoch": 0.7570621468926554,
+      "grad_norm": 2.266413927078247,
+      "learning_rate": 4.183479253901789e-06,
+      "loss": 1.4402,
+      "step": 737
+    },
+    {
+      "epoch": 0.7580893682588598,
+      "grad_norm": 2.2785308361053467,
+      "learning_rate": 4.181575942139323e-06,
+      "loss": 1.4874,
+      "step": 738
+    },
+    {
+      "epoch": 0.7591165896250642,
+      "grad_norm": 2.2710752487182617,
+      "learning_rate": 4.179672630376856e-06,
+      "loss": 1.4644,
+      "step": 739
+    },
+    {
+      "epoch": 0.7601438109912686,
+      "grad_norm": 2.2460498809814453,
+      "learning_rate": 4.177769318614389e-06,
+      "loss": 1.5267,
+      "step": 740
+    },
+    {
+      "epoch": 0.7601438109912686,
+      "eval_loss": 1.8187947273254395,
+      "eval_runtime": 37.9464,
+      "eval_samples_per_second": 26.353,
+      "eval_steps_per_second": 1.107,
+      "step": 740
+    },
+    {
+      "epoch": 0.7611710323574731,
+      "grad_norm": 2.3589370250701904,
+      "learning_rate": 4.175866006851923e-06,
+      "loss": 1.5115,
+      "step": 741
+    },
+    {
+      "epoch": 0.7621982537236774,
+      "grad_norm": 2.2818758487701416,
+      "learning_rate": 4.173962695089456e-06,
+      "loss": 1.4945,
+      "step": 742
+    },
+    {
+      "epoch": 0.7632254750898819,
+      "grad_norm": 2.338085889816284,
+      "learning_rate": 4.1720593833269894e-06,
+      "loss": 1.5076,
+      "step": 743
+    },
+    {
+      "epoch": 0.7642526964560863,
+      "grad_norm": 2.3874809741973877,
+      "learning_rate": 4.170156071564523e-06,
+      "loss": 1.5665,
+      "step": 744
+    },
+    {
+      "epoch": 0.7652799178222907,
+      "grad_norm": 2.2418739795684814,
+      "learning_rate": 4.168252759802056e-06,
+      "loss": 1.4827,
+      "step": 745
+    },
+    {
+      "epoch": 0.7663071391884951,
+      "grad_norm": 2.441315174102783,
+      "learning_rate": 4.1663494480395895e-06,
+      "loss": 1.5171,
+      "step": 746
+    },
+    {
+      "epoch": 0.7673343605546995,
+      "grad_norm": 2.3201963901519775,
+      "learning_rate": 4.164446136277123e-06,
+      "loss": 1.5153,
+      "step": 747
+    },
+    {
+      "epoch": 0.768361581920904,
+      "grad_norm": 2.3395135402679443,
+      "learning_rate": 4.162542824514656e-06,
+      "loss": 1.4911,
+      "step": 748
+    },
+    {
+      "epoch": 0.7693888032871083,
+      "grad_norm": 2.3455941677093506,
+      "learning_rate": 4.16063951275219e-06,
+      "loss": 1.4493,
+      "step": 749
+    },
+    {
+      "epoch": 0.7704160246533128,
+      "grad_norm": 2.3737382888793945,
+      "learning_rate": 4.158736200989722e-06,
+      "loss": 1.4893,
+      "step": 750
+    },
+    {
+      "epoch": 0.7704160246533128,
+      "eval_loss": 1.8153141736984253,
+      "eval_runtime": 37.9216,
+      "eval_samples_per_second": 26.37,
+      "eval_steps_per_second": 1.108,
+      "step": 750
+    },
+    {
+      "epoch": 0.7714432460195172,
+      "grad_norm": 2.503803014755249,
+      "learning_rate": 4.1568328892272555e-06,
+      "loss": 1.499,
+      "step": 751
+    },
+    {
+      "epoch": 0.7724704673857217,
+      "grad_norm": 2.469083547592163,
+      "learning_rate": 4.154929577464789e-06,
+      "loss": 1.4984,
+      "step": 752
+    },
+    {
+      "epoch": 0.773497688751926,
+      "grad_norm": 2.387991428375244,
+      "learning_rate": 4.153026265702322e-06,
+      "loss": 1.4283,
+      "step": 753
+    },
+    {
+      "epoch": 0.7745249101181304,
+      "grad_norm": 2.3355393409729004,
+      "learning_rate": 4.1511229539398556e-06,
+      "loss": 1.4869,
+      "step": 754
+    },
+    {
+      "epoch": 0.7755521314843349,
+      "grad_norm": 2.629345417022705,
+      "learning_rate": 4.149219642177389e-06,
+      "loss": 1.4885,
+      "step": 755
+    },
+    {
+      "epoch": 0.7765793528505393,
+      "grad_norm": 2.458944797515869,
+      "learning_rate": 4.147316330414922e-06,
+      "loss": 1.415,
+      "step": 756
+    },
+    {
+      "epoch": 0.7776065742167437,
+      "grad_norm": 2.40051531791687,
+      "learning_rate": 4.145413018652456e-06,
+      "loss": 1.5129,
+      "step": 757
+    },
+    {
+      "epoch": 0.7786337955829481,
+      "grad_norm": 2.5249011516571045,
+      "learning_rate": 4.143509706889989e-06,
+      "loss": 1.5002,
+      "step": 758
+    },
+    {
+      "epoch": 0.7796610169491526,
+      "grad_norm": 2.413422107696533,
+      "learning_rate": 4.141606395127522e-06,
+      "loss": 1.4837,
+      "step": 759
+    },
+    {
+      "epoch": 0.7806882383153569,
+      "grad_norm": 2.3438029289245605,
+      "learning_rate": 4.139703083365056e-06,
+      "loss": 1.5213,
+      "step": 760
+    },
+    {
+      "epoch": 0.7806882383153569,
+      "eval_loss": 1.8037596940994263,
+      "eval_runtime": 37.9229,
+      "eval_samples_per_second": 26.369,
+      "eval_steps_per_second": 1.108,
+      "step": 760
+    },
+    {
+      "epoch": 0.7817154596815614,
+      "grad_norm": 2.326828718185425,
+      "learning_rate": 4.137799771602588e-06,
+      "loss": 1.5031,
+      "step": 761
+    },
+    {
+      "epoch": 0.7827426810477658,
+      "grad_norm": 2.4902493953704834,
+      "learning_rate": 4.135896459840122e-06,
+      "loss": 1.4602,
+      "step": 762
+    },
+    {
+      "epoch": 0.7837699024139703,
+      "grad_norm": 2.488004684448242,
+      "learning_rate": 4.133993148077656e-06,
+      "loss": 1.4764,
+      "step": 763
+    },
+    {
+      "epoch": 0.7847971237801746,
+      "grad_norm": 2.5515761375427246,
+      "learning_rate": 4.132089836315189e-06,
+      "loss": 1.4946,
+      "step": 764
+    },
+    {
+      "epoch": 0.785824345146379,
+      "grad_norm": 2.556278705596924,
+      "learning_rate": 4.1301865245527226e-06,
+      "loss": 1.4338,
+      "step": 765
+    },
+    {
+      "epoch": 0.7868515665125835,
+      "grad_norm": 2.3017969131469727,
+      "learning_rate": 4.128283212790256e-06,
+      "loss": 1.4053,
+      "step": 766
+    },
+    {
+      "epoch": 0.7878787878787878,
+      "grad_norm": 2.4371304512023926,
+      "learning_rate": 4.1263799010277884e-06,
+      "loss": 1.5238,
+      "step": 767
+    },
+    {
+      "epoch": 0.7889060092449923,
+      "grad_norm": 2.6256046295166016,
+      "learning_rate": 4.124476589265322e-06,
+      "loss": 1.4672,
+      "step": 768
+    },
+    {
+      "epoch": 0.7899332306111967,
+      "grad_norm": 2.3725669384002686,
+      "learning_rate": 4.122573277502855e-06,
+      "loss": 1.4677,
+      "step": 769
+    },
+    {
+      "epoch": 0.7909604519774012,
+      "grad_norm": 2.2928454875946045,
+      "learning_rate": 4.1206699657403885e-06,
+      "loss": 1.4927,
+      "step": 770
+    },
+    {
+      "epoch": 0.7909604519774012,
+      "eval_loss": 1.792726755142212,
+      "eval_runtime": 37.9273,
+      "eval_samples_per_second": 26.366,
+      "eval_steps_per_second": 1.107,
+      "step": 770
+    },
+    {
+      "epoch": 0.7919876733436055,
+      "grad_norm": 2.315674066543579,
+      "learning_rate": 4.118766653977922e-06,
+      "loss": 1.4632,
+      "step": 771
+    },
+    {
+      "epoch": 0.79301489470981,
+      "grad_norm": 2.418849468231201,
+      "learning_rate": 4.116863342215455e-06,
+      "loss": 1.446,
+      "step": 772
+    },
+    {
+      "epoch": 0.7940421160760144,
+      "grad_norm": 2.3209879398345947,
+      "learning_rate": 4.114960030452989e-06,
+      "loss": 1.4755,
+      "step": 773
+    },
+    {
+      "epoch": 0.7950693374422187,
+      "grad_norm": 2.2322115898132324,
+      "learning_rate": 4.113056718690522e-06,
+      "loss": 1.4149,
+      "step": 774
+    },
+    {
+      "epoch": 0.7960965588084232,
+      "grad_norm": 2.6517558097839355,
+      "learning_rate": 4.111153406928055e-06,
+      "loss": 1.5323,
+      "step": 775
+    },
+    {
+      "epoch": 0.7971237801746276,
+      "grad_norm": 2.4602737426757812,
+      "learning_rate": 4.109250095165589e-06,
+      "loss": 1.5179,
+      "step": 776
+    },
+    {
+      "epoch": 0.7981510015408321,
+      "grad_norm": 2.4015536308288574,
+      "learning_rate": 4.107346783403122e-06,
+      "loss": 1.4711,
+      "step": 777
+    },
+    {
+      "epoch": 0.7991782229070364,
+      "grad_norm": 2.4367337226867676,
+      "learning_rate": 4.105443471640655e-06,
+      "loss": 1.4425,
+      "step": 778
+    },
+    {
+      "epoch": 0.8002054442732409,
+      "grad_norm": 2.5161020755767822,
+      "learning_rate": 4.103540159878188e-06,
+      "loss": 1.4832,
+      "step": 779
+    },
+    {
+      "epoch": 0.8012326656394453,
+      "grad_norm": 2.483154773712158,
+      "learning_rate": 4.101636848115721e-06,
+      "loss": 1.4508,
+      "step": 780
+    },
+    {
+      "epoch": 0.8012326656394453,
+      "eval_loss": 1.792397379875183,
+      "eval_runtime": 37.8937,
+      "eval_samples_per_second": 26.39,
+      "eval_steps_per_second": 1.108,
+      "step": 780
+    },
+    {
+      "epoch": 0.8022598870056498,
+      "grad_norm": 2.4649806022644043,
+      "learning_rate": 4.099733536353255e-06,
+      "loss": 1.4919,
+      "step": 781
+    },
+    {
+      "epoch": 0.8032871083718541,
+      "grad_norm": 2.578634262084961,
+      "learning_rate": 4.097830224590788e-06,
+      "loss": 1.565,
+      "step": 782
+    },
+    {
+      "epoch": 0.8043143297380585,
+      "grad_norm": 2.444427967071533,
+      "learning_rate": 4.095926912828322e-06,
+      "loss": 1.5309,
+      "step": 783
+    },
+    {
+      "epoch": 0.805341551104263,
+      "grad_norm": 2.32051420211792,
+      "learning_rate": 4.094023601065855e-06,
+      "loss": 1.4777,
+      "step": 784
+    },
+    {
+      "epoch": 0.8063687724704673,
+      "grad_norm": 2.3207180500030518,
+      "learning_rate": 4.092120289303388e-06,
+      "loss": 1.4863,
+      "step": 785
+    },
+    {
+      "epoch": 0.8073959938366718,
+      "grad_norm": 2.1404120922088623,
+      "learning_rate": 4.0902169775409215e-06,
+      "loss": 1.437,
+      "step": 786
+    },
+    {
+      "epoch": 0.8084232152028762,
+      "grad_norm": 2.2980856895446777,
+      "learning_rate": 4.088313665778455e-06,
+      "loss": 1.4571,
+      "step": 787
+    },
+    {
+      "epoch": 0.8094504365690807,
+      "grad_norm": 2.2555243968963623,
+      "learning_rate": 4.086410354015988e-06,
+      "loss": 1.4451,
+      "step": 788
+    },
+    {
+      "epoch": 0.810477657935285,
+      "grad_norm": 2.3837077617645264,
+      "learning_rate": 4.0845070422535216e-06,
+      "loss": 1.4641,
+      "step": 789
+    },
+    {
+      "epoch": 0.8115048793014895,
+      "grad_norm": 2.5199060440063477,
+      "learning_rate": 4.082603730491055e-06,
+      "loss": 1.5048,
+      "step": 790
+    },
+    {
+      "epoch": 0.8115048793014895,
+      "eval_loss": 1.8019492626190186,
+      "eval_runtime": 37.8805,
+      "eval_samples_per_second": 26.399,
+      "eval_steps_per_second": 1.109,
+      "step": 790
+    },
+    {
+      "epoch": 0.8125321006676939,
+      "grad_norm": 2.4010138511657715,
+      "learning_rate": 4.080700418728588e-06,
+      "loss": 1.4881,
+      "step": 791
+    },
+    {
+      "epoch": 0.8135593220338984,
+      "grad_norm": 2.342094898223877,
+      "learning_rate": 4.078797106966122e-06,
+      "loss": 1.3792,
+      "step": 792
+    },
+    {
+      "epoch": 0.8145865434001027,
+      "grad_norm": 2.276772975921631,
+      "learning_rate": 4.076893795203655e-06,
+      "loss": 1.4539,
+      "step": 793
+    },
+    {
+      "epoch": 0.8156137647663071,
+      "grad_norm": 2.569275140762329,
+      "learning_rate": 4.074990483441188e-06,
+      "loss": 1.5059,
+      "step": 794
+    },
+    {
+      "epoch": 0.8166409861325116,
+      "grad_norm": 2.345761299133301,
+      "learning_rate": 4.073087171678721e-06,
+      "loss": 1.494,
+      "step": 795
+    },
+    {
+      "epoch": 0.8176682074987159,
+      "grad_norm": 2.4425504207611084,
+      "learning_rate": 4.071183859916254e-06,
+      "loss": 1.471,
+      "step": 796
+    },
+    {
+      "epoch": 0.8186954288649204,
+      "grad_norm": 2.5758635997772217,
+      "learning_rate": 4.069280548153788e-06,
+      "loss": 1.4378,
+      "step": 797
+    },
+    {
+      "epoch": 0.8197226502311248,
+      "grad_norm": 2.4320778846740723,
+      "learning_rate": 4.067377236391321e-06,
+      "loss": 1.474,
+      "step": 798
+    },
+    {
+      "epoch": 0.8207498715973293,
+      "grad_norm": 2.466978073120117,
+      "learning_rate": 4.065473924628854e-06,
+      "loss": 1.4997,
+      "step": 799
+    },
+    {
+      "epoch": 0.8217770929635336,
+      "grad_norm": 2.462489128112793,
+      "learning_rate": 4.063570612866388e-06,
+      "loss": 1.5,
+      "step": 800
+    },
+    {
+      "epoch": 0.8217770929635336,
+      "eval_loss": 1.783312201499939,
+      "eval_runtime": 37.8832,
+      "eval_samples_per_second": 26.397,
+      "eval_steps_per_second": 1.109,
+      "step": 800
+    },
+    {
+      "epoch": 0.8228043143297381,
+      "grad_norm": 2.367455005645752,
+      "learning_rate": 4.061667301103921e-06,
+      "loss": 1.4658,
+      "step": 801
+    },
+    {
+      "epoch": 0.8238315356959425,
+      "grad_norm": 2.292250394821167,
+      "learning_rate": 4.0597639893414544e-06,
+      "loss": 1.476,
+      "step": 802
+    },
+    {
+      "epoch": 0.8248587570621468,
+      "grad_norm": 2.3460025787353516,
+      "learning_rate": 4.057860677578988e-06,
+      "loss": 1.4932,
+      "step": 803
+    },
+    {
+      "epoch": 0.8258859784283513,
+      "grad_norm": 2.3708906173706055,
+      "learning_rate": 4.055957365816521e-06,
+      "loss": 1.4396,
+      "step": 804
+    },
+    {
+      "epoch": 0.8269131997945557,
+      "grad_norm": 2.353513479232788,
+      "learning_rate": 4.0540540540540545e-06,
+      "loss": 1.4285,
+      "step": 805
+    },
+    {
+      "epoch": 0.8279404211607602,
+      "grad_norm": 2.319577932357788,
+      "learning_rate": 4.052150742291588e-06,
+      "loss": 1.4326,
+      "step": 806
+    },
+    {
+      "epoch": 0.8289676425269645,
+      "grad_norm": 2.450364112854004,
+      "learning_rate": 4.050247430529121e-06,
+      "loss": 1.5109,
+      "step": 807
+    },
+    {
+      "epoch": 0.829994863893169,
+      "grad_norm": 2.393907070159912,
+      "learning_rate": 4.048344118766655e-06,
+      "loss": 1.4948,
+      "step": 808
+    },
+    {
+      "epoch": 0.8310220852593734,
+      "grad_norm": 2.336002826690674,
+      "learning_rate": 4.046440807004188e-06,
+      "loss": 1.4851,
+      "step": 809
+    },
+    {
+      "epoch": 0.8320493066255779,
+      "grad_norm": 2.281139850616455,
+      "learning_rate": 4.044537495241721e-06,
+      "loss": 1.4445,
+      "step": 810
+    },
+    {
+      "epoch": 0.8320493066255779,
+      "eval_loss": 1.784231185913086,
+      "eval_runtime": 37.6614,
+      "eval_samples_per_second": 26.552,
+      "eval_steps_per_second": 1.115,
+      "step": 810
+    },
+    {
+      "epoch": 0.8330765279917822,
+      "grad_norm": 2.4848175048828125,
+      "learning_rate": 4.042634183479255e-06,
+      "loss": 1.4982,
+      "step": 811
+    },
+    {
+      "epoch": 0.8341037493579867,
+      "grad_norm": 2.263347864151001,
+      "learning_rate": 4.040730871716787e-06,
+      "loss": 1.4795,
+      "step": 812
+    },
+    {
+      "epoch": 0.8351309707241911,
+      "grad_norm": 2.394005060195923,
+      "learning_rate": 4.038827559954321e-06,
+      "loss": 1.5231,
+      "step": 813
+    },
+    {
+      "epoch": 0.8361581920903954,
+      "grad_norm": 2.5051357746124268,
+      "learning_rate": 4.036924248191854e-06,
+      "loss": 1.4886,
+      "step": 814
+    },
+    {
+      "epoch": 0.8371854134565999,
+      "grad_norm": 2.3580374717712402,
+      "learning_rate": 4.035020936429387e-06,
+      "loss": 1.4713,
+      "step": 815
+    },
+    {
+      "epoch": 0.8382126348228043,
+      "grad_norm": 2.3179264068603516,
+      "learning_rate": 4.033117624666921e-06,
+      "loss": 1.4369,
+      "step": 816
+    },
+    {
+      "epoch": 0.8392398561890088,
+      "grad_norm": 2.269911289215088,
+      "learning_rate": 4.031214312904454e-06,
+      "loss": 1.4326,
+      "step": 817
+    },
+    {
+      "epoch": 0.8402670775552131,
+      "grad_norm": 2.3177807331085205,
+      "learning_rate": 4.029311001141987e-06,
+      "loss": 1.4096,
+      "step": 818
+    },
+    {
+      "epoch": 0.8412942989214176,
+      "grad_norm": 2.394324541091919,
+      "learning_rate": 4.027407689379521e-06,
+      "loss": 1.4647,
+      "step": 819
+    },
+    {
+      "epoch": 0.842321520287622,
+      "grad_norm": 2.288536548614502,
+      "learning_rate": 4.025504377617054e-06,
+      "loss": 1.4472,
+      "step": 820
+    },
+    {
+      "epoch": 0.842321520287622,
+      "eval_loss": 1.7811495065689087,
+      "eval_runtime": 37.8132,
+      "eval_samples_per_second": 26.446,
+      "eval_steps_per_second": 1.111,
+      "step": 820
+    },
+    {
+      "epoch": 0.8433487416538265,
+      "grad_norm": 2.376574754714966,
+      "learning_rate": 4.0236010658545875e-06,
+      "loss": 1.4918,
+      "step": 821
+    },
+    {
+      "epoch": 0.8443759630200308,
+      "grad_norm": 2.5850327014923096,
+      "learning_rate": 4.02169775409212e-06,
+      "loss": 1.5342,
+      "step": 822
+    },
+    {
+      "epoch": 0.8454031843862352,
+      "grad_norm": 2.247206211090088,
+      "learning_rate": 4.019794442329653e-06,
+      "loss": 1.4177,
+      "step": 823
+    },
+    {
+      "epoch": 0.8464304057524397,
+      "grad_norm": 2.4411795139312744,
+      "learning_rate": 4.0178911305671876e-06,
+      "loss": 1.4992,
+      "step": 824
+    },
+    {
+      "epoch": 0.847457627118644,
+      "grad_norm": 2.3377397060394287,
+      "learning_rate": 4.015987818804721e-06,
+      "loss": 1.4359,
+      "step": 825
+    },
+    {
+      "epoch": 0.8484848484848485,
+      "grad_norm": 2.329857349395752,
+      "learning_rate": 4.014084507042254e-06,
+      "loss": 1.4096,
+      "step": 826
+    },
+    {
+      "epoch": 0.8495120698510529,
+      "grad_norm": 2.563685178756714,
+      "learning_rate": 4.012181195279788e-06,
+      "loss": 1.5174,
+      "step": 827
+    },
+    {
+      "epoch": 0.8505392912172574,
+      "grad_norm": 2.4521079063415527,
+      "learning_rate": 4.010277883517321e-06,
+      "loss": 1.5028,
+      "step": 828
+    },
+    {
+      "epoch": 0.8515665125834617,
+      "grad_norm": 2.4091477394104004,
+      "learning_rate": 4.0083745717548535e-06,
+      "loss": 1.4615,
+      "step": 829
+    },
+    {
+      "epoch": 0.8525937339496662,
+      "grad_norm": 2.4010677337646484,
+      "learning_rate": 4.006471259992387e-06,
+      "loss": 1.4834,
+      "step": 830
+    },
+    {
+      "epoch": 0.8525937339496662,
+      "eval_loss": 1.7797678709030151,
+      "eval_runtime": 37.9378,
+      "eval_samples_per_second": 26.359,
+      "eval_steps_per_second": 1.107,
+      "step": 830
+    },
+    {
+      "epoch": 0.8536209553158706,
+      "grad_norm": 2.3712635040283203,
+      "learning_rate": 4.00456794822992e-06,
+      "loss": 1.5196,
+      "step": 831
+    },
+    {
+      "epoch": 0.8546481766820749,
+      "grad_norm": 2.4096601009368896,
+      "learning_rate": 4.002664636467454e-06,
+      "loss": 1.4769,
+      "step": 832
+    },
+    {
+      "epoch": 0.8556753980482794,
+      "grad_norm": 2.408466100692749,
+      "learning_rate": 4.000761324704987e-06,
+      "loss": 1.4268,
+      "step": 833
+    },
+    {
+      "epoch": 0.8567026194144838,
+      "grad_norm": 2.6857047080993652,
+      "learning_rate": 3.99885801294252e-06,
+      "loss": 1.492,
+      "step": 834
+    },
+    {
+      "epoch": 0.8577298407806883,
+      "grad_norm": 2.297550916671753,
+      "learning_rate": 3.996954701180054e-06,
+      "loss": 1.4184,
+      "step": 835
+    },
+    {
+      "epoch": 0.8587570621468926,
+      "grad_norm": 2.423224925994873,
+      "learning_rate": 3.995051389417587e-06,
+      "loss": 1.4337,
+      "step": 836
+    },
+    {
+      "epoch": 0.8597842835130971,
+      "grad_norm": 2.260493040084839,
+      "learning_rate": 3.9931480776551204e-06,
+      "loss": 1.5267,
+      "step": 837
+    },
+    {
+      "epoch": 0.8608115048793015,
+      "grad_norm": 2.3803012371063232,
+      "learning_rate": 3.991244765892654e-06,
+      "loss": 1.4426,
+      "step": 838
+    },
+    {
+      "epoch": 0.861838726245506,
+      "grad_norm": 2.1425721645355225,
+      "learning_rate": 3.989341454130186e-06,
+      "loss": 1.4148,
+      "step": 839
+    },
+    {
+      "epoch": 0.8628659476117103,
+      "grad_norm": 2.316624641418457,
+      "learning_rate": 3.98743814236772e-06,
+      "loss": 1.4544,
+      "step": 840
+    },
+    {
+      "epoch": 0.8628659476117103,
+      "eval_loss": 1.7824597358703613,
+      "eval_runtime": 37.9676,
+      "eval_samples_per_second": 26.338,
+      "eval_steps_per_second": 1.106,
+      "step": 840
+    },
+    {
+      "epoch": 0.8638931689779148,
+      "grad_norm": 2.3723156452178955,
+      "learning_rate": 3.985534830605253e-06,
+      "loss": 1.473,
+      "step": 841
+    },
+    {
+      "epoch": 0.8649203903441192,
+      "grad_norm": 2.5167760848999023,
+      "learning_rate": 3.983631518842786e-06,
+      "loss": 1.4919,
+      "step": 842
+    },
+    {
+      "epoch": 0.8659476117103235,
+      "grad_norm": 2.3018670082092285,
+      "learning_rate": 3.98172820708032e-06,
+      "loss": 1.447,
+      "step": 843
+    },
+    {
+      "epoch": 0.866974833076528,
+      "grad_norm": 2.420483112335205,
+      "learning_rate": 3.979824895317854e-06,
+      "loss": 1.4159,
+      "step": 844
+    },
+    {
+      "epoch": 0.8680020544427324,
+      "grad_norm": 2.39443039894104,
+      "learning_rate": 3.977921583555387e-06,
+      "loss": 1.4984,
+      "step": 845
+    },
+    {
+      "epoch": 0.8690292758089369,
+      "grad_norm": 2.2953920364379883,
+      "learning_rate": 3.97601827179292e-06,
+      "loss": 1.4297,
+      "step": 846
+    },
+    {
+      "epoch": 0.8700564971751412,
+      "grad_norm": 2.2919528484344482,
+      "learning_rate": 3.974114960030453e-06,
+      "loss": 1.4728,
+      "step": 847
+    },
+    {
+      "epoch": 0.8710837185413457,
+      "grad_norm": 2.327914237976074,
+      "learning_rate": 3.9722116482679866e-06,
+      "loss": 1.4584,
+      "step": 848
+    },
+    {
+      "epoch": 0.8721109399075501,
+      "grad_norm": 2.305720567703247,
+      "learning_rate": 3.97030833650552e-06,
+      "loss": 1.4688,
+      "step": 849
+    },
+    {
+      "epoch": 0.8731381612737545,
+      "grad_norm": 2.2969236373901367,
+      "learning_rate": 3.968405024743053e-06,
+      "loss": 1.3972,
+      "step": 850
+    },
+    {
+      "epoch": 0.8731381612737545,
+      "eval_loss": 1.775634765625,
+      "eval_runtime": 37.9916,
+      "eval_samples_per_second": 26.322,
+      "eval_steps_per_second": 1.106,
+      "step": 850
+    },
+    {
+      "epoch": 0.8741653826399589,
+      "grad_norm": 2.314387083053589,
+      "learning_rate": 3.966501712980587e-06,
+      "loss": 1.4476,
+      "step": 851
+    },
+    {
+      "epoch": 0.8751926040061633,
+      "grad_norm": 2.3271584510803223,
+      "learning_rate": 3.96459840121812e-06,
+      "loss": 1.4615,
+      "step": 852
+    },
+    {
+      "epoch": 0.8762198253723678,
+      "grad_norm": 2.494795322418213,
+      "learning_rate": 3.962695089455653e-06,
+      "loss": 1.5062,
+      "step": 853
+    },
+    {
+      "epoch": 0.8772470467385721,
+      "grad_norm": 2.4132444858551025,
+      "learning_rate": 3.960791777693187e-06,
+      "loss": 1.4861,
+      "step": 854
+    },
+    {
+      "epoch": 0.8782742681047766,
+      "grad_norm": 2.3869900703430176,
+      "learning_rate": 3.95888846593072e-06,
+      "loss": 1.4923,
+      "step": 855
+    },
+    {
+      "epoch": 0.879301489470981,
+      "grad_norm": 2.5167367458343506,
+      "learning_rate": 3.956985154168253e-06,
+      "loss": 1.5137,
+      "step": 856
+    },
+    {
+      "epoch": 0.8803287108371854,
+      "grad_norm": 2.510361671447754,
+      "learning_rate": 3.955081842405786e-06,
+      "loss": 1.5414,
+      "step": 857
+    },
+    {
+      "epoch": 0.8813559322033898,
+      "grad_norm": 2.554427146911621,
+      "learning_rate": 3.953178530643319e-06,
+      "loss": 1.5569,
+      "step": 858
+    },
+    {
+      "epoch": 0.8823831535695943,
+      "grad_norm": 2.306509256362915,
+      "learning_rate": 3.951275218880853e-06,
+      "loss": 1.3982,
+      "step": 859
+    },
+    {
+      "epoch": 0.8834103749357987,
+      "grad_norm": 2.3805363178253174,
+      "learning_rate": 3.949371907118386e-06,
+      "loss": 1.4977,
+      "step": 860
+    },
+    {
+      "epoch": 0.8834103749357987,
+      "eval_loss": 1.7882694005966187,
+      "eval_runtime": 38.0072,
+      "eval_samples_per_second": 26.311,
+      "eval_steps_per_second": 1.105,
+      "step": 860
+    },
+    {
+      "epoch": 0.884437596302003,
+      "grad_norm": 2.5413448810577393,
+      "learning_rate": 3.9474685953559194e-06,
+      "loss": 1.472,
+      "step": 861
+    },
+    {
+      "epoch": 0.8854648176682075,
+      "grad_norm": 2.484834909439087,
+      "learning_rate": 3.945565283593453e-06,
+      "loss": 1.4597,
+      "step": 862
+    },
+    {
+      "epoch": 0.8864920390344119,
+      "grad_norm": 2.6060585975646973,
+      "learning_rate": 3.943661971830986e-06,
+      "loss": 1.509,
+      "step": 863
+    },
+    {
+      "epoch": 0.8875192604006163,
+      "grad_norm": 2.6536927223205566,
+      "learning_rate": 3.9417586600685195e-06,
+      "loss": 1.4836,
+      "step": 864
+    },
+    {
+      "epoch": 0.8885464817668207,
+      "grad_norm": 2.5198960304260254,
+      "learning_rate": 3.939855348306053e-06,
+      "loss": 1.486,
+      "step": 865
+    },
+    {
+      "epoch": 0.8895737031330252,
+      "grad_norm": 2.335204839706421,
+      "learning_rate": 3.937952036543586e-06,
+      "loss": 1.4701,
+      "step": 866
+    },
+    {
+      "epoch": 0.8906009244992296,
+      "grad_norm": 2.7058098316192627,
+      "learning_rate": 3.93604872478112e-06,
+      "loss": 1.4798,
+      "step": 867
+    },
+    {
+      "epoch": 0.891628145865434,
+      "grad_norm": 2.4798922538757324,
+      "learning_rate": 3.934145413018653e-06,
+      "loss": 1.4364,
+      "step": 868
+    },
+    {
+      "epoch": 0.8926553672316384,
+      "grad_norm": 2.3090760707855225,
+      "learning_rate": 3.932242101256186e-06,
+      "loss": 1.4405,
+      "step": 869
+    },
+    {
+      "epoch": 0.8936825885978429,
+      "grad_norm": 2.4251503944396973,
+      "learning_rate": 3.93033878949372e-06,
+      "loss": 1.4113,
+      "step": 870
+    },
+    {
+      "epoch": 0.8936825885978429,
+      "eval_loss": 1.794334888458252,
+      "eval_runtime": 38.0197,
+      "eval_samples_per_second": 26.302,
+      "eval_steps_per_second": 1.105,
+      "step": 870
+    },
+    {
+      "epoch": 0.8947098099640473,
+      "grad_norm": 2.594945192337036,
+      "learning_rate": 3.928435477731253e-06,
+      "loss": 1.502,
+      "step": 871
+    },
+    {
+      "epoch": 0.8957370313302516,
+      "grad_norm": 2.4224843978881836,
+      "learning_rate": 3.9265321659687864e-06,
+      "loss": 1.5005,
+      "step": 872
+    },
+    {
+      "epoch": 0.8967642526964561,
+      "grad_norm": 2.3968186378479004,
+      "learning_rate": 3.924628854206319e-06,
+      "loss": 1.481,
+      "step": 873
+    },
+    {
+      "epoch": 0.8977914740626605,
+      "grad_norm": 2.4923524856567383,
+      "learning_rate": 3.922725542443852e-06,
+      "loss": 1.4259,
+      "step": 874
+    },
+    {
+      "epoch": 0.8988186954288649,
+      "grad_norm": 2.5386104583740234,
+      "learning_rate": 3.920822230681386e-06,
+      "loss": 1.4497,
+      "step": 875
+    },
+    {
+      "epoch": 0.8998459167950693,
+      "grad_norm": 2.4014909267425537,
+      "learning_rate": 3.918918918918919e-06,
+      "loss": 1.4816,
+      "step": 876
+    },
+    {
+      "epoch": 0.9008731381612738,
+      "grad_norm": 2.4019317626953125,
+      "learning_rate": 3.917015607156452e-06,
+      "loss": 1.4669,
+      "step": 877
+    },
+    {
+      "epoch": 0.9019003595274782,
+      "grad_norm": 2.5408380031585693,
+      "learning_rate": 3.915112295393986e-06,
+      "loss": 1.5343,
+      "step": 878
+    },
+    {
+      "epoch": 0.9029275808936826,
+      "grad_norm": 2.364053964614868,
+      "learning_rate": 3.913208983631519e-06,
+      "loss": 1.5229,
+      "step": 879
+    },
+    {
+      "epoch": 0.903954802259887,
+      "grad_norm": 2.307412624359131,
+      "learning_rate": 3.9113056718690525e-06,
+      "loss": 1.4444,
+      "step": 880
+    },
+    {
+      "epoch": 0.903954802259887,
+      "eval_loss": 1.7910975217819214,
+      "eval_runtime": 38.0659,
+      "eval_samples_per_second": 26.27,
+      "eval_steps_per_second": 1.103,
+      "step": 880
+    },
+    {
+      "epoch": 0.9049820236260914,
+      "grad_norm": 2.496135950088501,
+      "learning_rate": 3.909402360106586e-06,
+      "loss": 1.4853,
+      "step": 881
+    },
+    {
+      "epoch": 0.9060092449922958,
+      "grad_norm": 2.4825589656829834,
+      "learning_rate": 3.907499048344119e-06,
+      "loss": 1.4886,
+      "step": 882
+    },
+    {
+      "epoch": 0.9070364663585002,
+      "grad_norm": 2.3458824157714844,
+      "learning_rate": 3.905595736581652e-06,
+      "loss": 1.4882,
+      "step": 883
+    },
+    {
+      "epoch": 0.9080636877247047,
+      "grad_norm": 2.2699403762817383,
+      "learning_rate": 3.903692424819186e-06,
+      "loss": 1.4417,
+      "step": 884
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 2.348593235015869,
+      "learning_rate": 3.901789113056719e-06,
+      "loss": 1.4991,
+      "step": 885
+    },
+    {
+      "epoch": 0.9101181304571135,
+      "grad_norm": 2.3292129039764404,
+      "learning_rate": 3.899885801294253e-06,
+      "loss": 1.4752,
+      "step": 886
+    },
+    {
+      "epoch": 0.9111453518233179,
+      "grad_norm": 2.487183094024658,
+      "learning_rate": 3.897982489531786e-06,
+      "loss": 1.5317,
+      "step": 887
+    },
+    {
+      "epoch": 0.9121725731895224,
+      "grad_norm": 2.3300845623016357,
+      "learning_rate": 3.896079177769319e-06,
+      "loss": 1.4964,
+      "step": 888
+    },
+    {
+      "epoch": 0.9131997945557268,
+      "grad_norm": 2.3292312622070312,
+      "learning_rate": 3.894175866006853e-06,
+      "loss": 1.5221,
+      "step": 889
+    },
+    {
+      "epoch": 0.9142270159219312,
+      "grad_norm": 2.3556439876556396,
+      "learning_rate": 3.892272554244385e-06,
+      "loss": 1.4471,
+      "step": 890
+    },
+    {
+      "epoch": 0.9142270159219312,
+      "eval_loss": 1.7807039022445679,
+      "eval_runtime": 38.0614,
+      "eval_samples_per_second": 26.273,
+      "eval_steps_per_second": 1.103,
+      "step": 890
+    },
+    {
+      "epoch": 0.9152542372881356,
+      "grad_norm": 2.3302760124206543,
+      "learning_rate": 3.890369242481919e-06,
+      "loss": 1.5179,
+      "step": 891
+    },
+    {
+      "epoch": 0.91628145865434,
+      "grad_norm": 2.327072858810425,
+      "learning_rate": 3.888465930719452e-06,
+      "loss": 1.4474,
+      "step": 892
+    },
+    {
+      "epoch": 0.9173086800205444,
+      "grad_norm": 2.2773101329803467,
+      "learning_rate": 3.886562618956985e-06,
+      "loss": 1.4783,
+      "step": 893
+    },
+    {
+      "epoch": 0.9183359013867488,
+      "grad_norm": 2.2717974185943604,
+      "learning_rate": 3.884659307194519e-06,
+      "loss": 1.4062,
+      "step": 894
+    },
+    {
+      "epoch": 0.9193631227529533,
+      "grad_norm": 2.3350367546081543,
+      "learning_rate": 3.882755995432052e-06,
+      "loss": 1.483,
+      "step": 895
+    },
+    {
+      "epoch": 0.9203903441191577,
+      "grad_norm": 2.302811622619629,
+      "learning_rate": 3.8808526836695854e-06,
+      "loss": 1.4407,
+      "step": 896
+    },
+    {
+      "epoch": 0.9214175654853621,
+      "grad_norm": 2.4031271934509277,
+      "learning_rate": 3.878949371907119e-06,
+      "loss": 1.4331,
+      "step": 897
+    },
+    {
+      "epoch": 0.9224447868515665,
+      "grad_norm": 2.4806716442108154,
+      "learning_rate": 3.877046060144652e-06,
+      "loss": 1.4341,
+      "step": 898
+    },
+    {
+      "epoch": 0.923472008217771,
+      "grad_norm": 2.4243357181549072,
+      "learning_rate": 3.8751427483821855e-06,
+      "loss": 1.4701,
+      "step": 899
+    },
+    {
+      "epoch": 0.9244992295839753,
+      "grad_norm": 2.4953269958496094,
+      "learning_rate": 3.873239436619718e-06,
+      "loss": 1.4287,
+      "step": 900
+    },
+    {
+      "epoch": 0.9244992295839753,
+      "eval_loss": 1.7872728109359741,
+      "eval_runtime": 38.0906,
+      "eval_samples_per_second": 26.253,
+      "eval_steps_per_second": 1.103,
+      "step": 900
+    },
+    {
+      "epoch": 0.9255264509501797,
+      "grad_norm": 2.307490587234497,
+      "learning_rate": 3.871336124857251e-06,
+      "loss": 1.4693,
+      "step": 901
+    },
+    {
+      "epoch": 0.9265536723163842,
+      "grad_norm": 2.4745755195617676,
+      "learning_rate": 3.869432813094785e-06,
+      "loss": 1.4779,
+      "step": 902
+    },
+    {
+      "epoch": 0.9275808936825886,
+      "grad_norm": 2.367715358734131,
+      "learning_rate": 3.867529501332318e-06,
+      "loss": 1.4771,
+      "step": 903
+    },
+    {
+      "epoch": 0.928608115048793,
+      "grad_norm": 2.388549566268921,
+      "learning_rate": 3.8656261895698515e-06,
+      "loss": 1.4648,
+      "step": 904
+    },
+    {
+      "epoch": 0.9296353364149974,
+      "grad_norm": 2.39432954788208,
+      "learning_rate": 3.863722877807386e-06,
+      "loss": 1.4932,
+      "step": 905
+    },
+    {
+      "epoch": 0.9306625577812019,
+      "grad_norm": 2.3921315670013428,
+      "learning_rate": 3.861819566044919e-06,
+      "loss": 1.5312,
+      "step": 906
+    },
+    {
+      "epoch": 0.9316897791474062,
+      "grad_norm": 2.400240659713745,
+      "learning_rate": 3.859916254282452e-06,
+      "loss": 1.4651,
+      "step": 907
+    },
+    {
+      "epoch": 0.9327170005136107,
+      "grad_norm": 2.394559860229492,
+      "learning_rate": 3.858012942519985e-06,
+      "loss": 1.4759,
+      "step": 908
+    },
+    {
+      "epoch": 0.9337442218798151,
+      "grad_norm": 2.3365273475646973,
+      "learning_rate": 3.856109630757518e-06,
+      "loss": 1.4502,
+      "step": 909
+    },
+    {
+      "epoch": 0.9347714432460195,
+      "grad_norm": 2.465691328048706,
+      "learning_rate": 3.854206318995052e-06,
+      "loss": 1.5036,
+      "step": 910
+    },
+    {
+      "epoch": 0.9347714432460195,
+      "eval_loss": 1.7682362794876099,
+      "eval_runtime": 37.6624,
+      "eval_samples_per_second": 26.552,
+      "eval_steps_per_second": 1.115,
+      "step": 910
+    },
+    {
+      "epoch": 0.9357986646122239,
+      "grad_norm": 2.3376758098602295,
+      "learning_rate": 3.852303007232585e-06,
+      "loss": 1.4342,
+      "step": 911
+    },
+    {
+      "epoch": 0.9368258859784283,
+      "grad_norm": 2.345855236053467,
+      "learning_rate": 3.850399695470118e-06,
+      "loss": 1.4444,
+      "step": 912
+    },
+    {
+      "epoch": 0.9378531073446328,
+      "grad_norm": 2.4444501399993896,
+      "learning_rate": 3.848496383707652e-06,
+      "loss": 1.4315,
+      "step": 913
+    },
+    {
+      "epoch": 0.9388803287108372,
+      "grad_norm": 2.4896373748779297,
+      "learning_rate": 3.846593071945185e-06,
+      "loss": 1.4228,
+      "step": 914
+    },
+    {
+      "epoch": 0.9399075500770416,
+      "grad_norm": 2.3970463275909424,
+      "learning_rate": 3.8446897601827185e-06,
+      "loss": 1.5283,
+      "step": 915
+    },
+    {
+      "epoch": 0.940934771443246,
+      "grad_norm": 2.5194883346557617,
+      "learning_rate": 3.842786448420252e-06,
+      "loss": 1.5445,
+      "step": 916
+    },
+    {
+      "epoch": 0.9419619928094505,
+      "grad_norm": 2.446465253829956,
+      "learning_rate": 3.840883136657785e-06,
+      "loss": 1.5248,
+      "step": 917
+    },
+    {
+      "epoch": 0.9429892141756548,
+      "grad_norm": 2.348477840423584,
+      "learning_rate": 3.838979824895318e-06,
+      "loss": 1.402,
+      "step": 918
+    },
+    {
+      "epoch": 0.9440164355418593,
+      "grad_norm": 2.4511539936065674,
+      "learning_rate": 3.837076513132851e-06,
+      "loss": 1.4893,
+      "step": 919
+    },
+    {
+      "epoch": 0.9450436569080637,
+      "grad_norm": 2.3801777362823486,
+      "learning_rate": 3.8351732013703844e-06,
+      "loss": 1.4457,
+      "step": 920
+    },
+    {
+      "epoch": 0.9450436569080637,
+      "eval_loss": 1.7810813188552856,
+      "eval_runtime": 37.7647,
+      "eval_samples_per_second": 26.48,
+      "eval_steps_per_second": 1.112,
+      "step": 920
+    },
+    {
+      "epoch": 0.9460708782742681,
+      "grad_norm": 2.5260043144226074,
+      "learning_rate": 3.833269889607918e-06,
+      "loss": 1.4267,
+      "step": 921
+    },
+    {
+      "epoch": 0.9470980996404725,
+      "grad_norm": 2.270653009414673,
+      "learning_rate": 3.831366577845451e-06,
+      "loss": 1.4313,
+      "step": 922
+    },
+    {
+      "epoch": 0.9481253210066769,
+      "grad_norm": 2.5123250484466553,
+      "learning_rate": 3.8294632660829845e-06,
+      "loss": 1.4588,
+      "step": 923
+    },
+    {
+      "epoch": 0.9491525423728814,
+      "grad_norm": 2.4242913722991943,
+      "learning_rate": 3.827559954320518e-06,
+      "loss": 1.4659,
+      "step": 924
+    },
+    {
+      "epoch": 0.9501797637390857,
+      "grad_norm": 2.4337117671966553,
+      "learning_rate": 3.825656642558051e-06,
+      "loss": 1.4838,
+      "step": 925
+    },
+    {
+      "epoch": 0.9512069851052902,
+      "grad_norm": 2.3577871322631836,
+      "learning_rate": 3.823753330795585e-06,
+      "loss": 1.4674,
+      "step": 926
+    },
+    {
+      "epoch": 0.9522342064714946,
+      "grad_norm": 2.2574679851531982,
+      "learning_rate": 3.821850019033118e-06,
+      "loss": 1.4011,
+      "step": 927
+    },
+    {
+      "epoch": 0.9532614278376991,
+      "grad_norm": 2.38043475151062,
+      "learning_rate": 3.819946707270651e-06,
+      "loss": 1.459,
+      "step": 928
+    },
+    {
+      "epoch": 0.9542886492039034,
+      "grad_norm": 2.4249954223632812,
+      "learning_rate": 3.818043395508185e-06,
+      "loss": 1.4649,
+      "step": 929
+    },
+    {
+      "epoch": 0.9553158705701078,
+      "grad_norm": 2.4420788288116455,
+      "learning_rate": 3.816140083745718e-06,
+      "loss": 1.532,
+      "step": 930
+    },
+    {
+      "epoch": 0.9553158705701078,
+      "eval_loss": 1.7805145978927612,
+      "eval_runtime": 37.8606,
+      "eval_samples_per_second": 26.413,
+      "eval_steps_per_second": 1.109,
+      "step": 930
+    },
+    {
+      "epoch": 0.9563430919363123,
+      "grad_norm": 2.391042470932007,
+      "learning_rate": 3.8142367719832514e-06,
+      "loss": 1.4365,
+      "step": 931
+    },
+    {
+      "epoch": 0.9573703133025167,
+      "grad_norm": 2.462303876876831,
+      "learning_rate": 3.8123334602207844e-06,
+      "loss": 1.4367,
+      "step": 932
+    },
+    {
+      "epoch": 0.9583975346687211,
+      "grad_norm": 2.330714702606201,
+      "learning_rate": 3.8104301484583177e-06,
+      "loss": 1.427,
+      "step": 933
+    },
+    {
+      "epoch": 0.9594247560349255,
+      "grad_norm": 2.3475441932678223,
+      "learning_rate": 3.808526836695851e-06,
+      "loss": 1.4836,
+      "step": 934
+    },
+    {
+      "epoch": 0.96045197740113,
+      "grad_norm": 2.4546470642089844,
+      "learning_rate": 3.8066235249333845e-06,
+      "loss": 1.4601,
+      "step": 935
+    },
+    {
+      "epoch": 0.9614791987673343,
+      "grad_norm": 2.4604640007019043,
+      "learning_rate": 3.804720213170918e-06,
+      "loss": 1.4616,
+      "step": 936
+    },
+    {
+      "epoch": 0.9625064201335388,
+      "grad_norm": 2.3087422847747803,
+      "learning_rate": 3.8028169014084508e-06,
+      "loss": 1.4816,
+      "step": 937
+    },
+    {
+      "epoch": 0.9635336414997432,
+      "grad_norm": 2.375377893447876,
+      "learning_rate": 3.800913589645984e-06,
+      "loss": 1.471,
+      "step": 938
+    },
+    {
+      "epoch": 0.9645608628659477,
+      "grad_norm": 2.358741283416748,
+      "learning_rate": 3.7990102778835175e-06,
+      "loss": 1.4754,
+      "step": 939
+    },
+    {
+      "epoch": 0.965588084232152,
+      "grad_norm": 2.3990139961242676,
+      "learning_rate": 3.797106966121051e-06,
+      "loss": 1.534,
+      "step": 940
+    },
+    {
+      "epoch": 0.965588084232152,
+      "eval_loss": 1.7745739221572876,
+      "eval_runtime": 37.8883,
+      "eval_samples_per_second": 26.393,
+      "eval_steps_per_second": 1.109,
+      "step": 940
+    },
+    {
+      "epoch": 0.9666153055983564,
+      "grad_norm": 2.417396068572998,
+      "learning_rate": 3.7952036543585842e-06,
+      "loss": 1.4994,
+      "step": 941
+    },
+    {
+      "epoch": 0.9676425269645609,
+      "grad_norm": 2.323125123977661,
+      "learning_rate": 3.793300342596117e-06,
+      "loss": 1.5045,
+      "step": 942
+    },
+    {
+      "epoch": 0.9686697483307652,
+      "grad_norm": 2.4424829483032227,
+      "learning_rate": 3.7913970308336505e-06,
+      "loss": 1.4936,
+      "step": 943
+    },
+    {
+      "epoch": 0.9696969696969697,
+      "grad_norm": 2.2769429683685303,
+      "learning_rate": 3.789493719071184e-06,
+      "loss": 1.4453,
+      "step": 944
+    },
+    {
+      "epoch": 0.9707241910631741,
+      "grad_norm": 2.183701276779175,
+      "learning_rate": 3.7875904073087177e-06,
+      "loss": 1.4005,
+      "step": 945
+    },
+    {
+      "epoch": 0.9717514124293786,
+      "grad_norm": 3.4538655281066895,
+      "learning_rate": 3.785687095546251e-06,
+      "loss": 1.4521,
+      "step": 946
+    },
+    {
+      "epoch": 0.9727786337955829,
+      "grad_norm": 2.3735523223876953,
+      "learning_rate": 3.7837837837837844e-06,
+      "loss": 1.4328,
+      "step": 947
+    },
+    {
+      "epoch": 0.9738058551617874,
+      "grad_norm": 2.6214165687561035,
+      "learning_rate": 3.7818804720213178e-06,
+      "loss": 1.4004,
+      "step": 948
+    },
+    {
+      "epoch": 0.9748330765279918,
+      "grad_norm": 2.365091562271118,
+      "learning_rate": 3.7799771602588507e-06,
+      "loss": 1.4006,
+      "step": 949
+    },
+    {
+      "epoch": 0.9758602978941961,
+      "grad_norm": 2.1926634311676025,
+      "learning_rate": 3.778073848496384e-06,
+      "loss": 1.4587,
+      "step": 950
+    },
+    {
+      "epoch": 0.9758602978941961,
+      "eval_loss": 1.767502784729004,
+      "eval_runtime": 37.9287,
+      "eval_samples_per_second": 26.365,
+      "eval_steps_per_second": 1.107,
+      "step": 950
+    },
+    {
+      "epoch": 0.9768875192604006,
+      "grad_norm": 2.3047192096710205,
+      "learning_rate": 3.7761705367339174e-06,
+      "loss": 1.4826,
+      "step": 951
+    },
+    {
+      "epoch": 0.977914740626605,
+      "grad_norm": 2.3306028842926025,
+      "learning_rate": 3.7742672249714508e-06,
+      "loss": 1.4914,
+      "step": 952
+    },
+    {
+      "epoch": 0.9789419619928095,
+      "grad_norm": 2.3409693241119385,
+      "learning_rate": 3.772363913208984e-06,
+      "loss": 1.4556,
+      "step": 953
+    },
+    {
+      "epoch": 0.9799691833590138,
+      "grad_norm": 2.3337934017181396,
+      "learning_rate": 3.770460601446517e-06,
+      "loss": 1.4751,
+      "step": 954
+    },
+    {
+      "epoch": 0.9809964047252183,
+      "grad_norm": 2.3354830741882324,
+      "learning_rate": 3.7685572896840504e-06,
+      "loss": 1.4176,
+      "step": 955
+    },
+    {
+      "epoch": 0.9820236260914227,
+      "grad_norm": 2.4049782752990723,
+      "learning_rate": 3.766653977921584e-06,
+      "loss": 1.3715,
+      "step": 956
+    },
+    {
+      "epoch": 0.9830508474576272,
+      "grad_norm": 2.5127065181732178,
+      "learning_rate": 3.764750666159117e-06,
+      "loss": 1.468,
+      "step": 957
+    },
+    {
+      "epoch": 0.9840780688238315,
+      "grad_norm": 2.5576820373535156,
+      "learning_rate": 3.7628473543966505e-06,
+      "loss": 1.4237,
+      "step": 958
+    },
+    {
+      "epoch": 0.9851052901900359,
+      "grad_norm": 2.394098997116089,
+      "learning_rate": 3.7609440426341835e-06,
+      "loss": 1.5126,
+      "step": 959
+    },
+    {
+      "epoch": 0.9861325115562404,
+      "grad_norm": 2.3358702659606934,
+      "learning_rate": 3.759040730871717e-06,
+      "loss": 1.4109,
+      "step": 960
+    },
+    {
+      "epoch": 0.9861325115562404,
+      "eval_loss": 1.7665778398513794,
+      "eval_runtime": 37.9028,
+      "eval_samples_per_second": 26.383,
+      "eval_steps_per_second": 1.108,
+      "step": 960
+    },
+    {
+      "epoch": 0.9871597329224447,
+      "grad_norm": 2.4213271141052246,
+      "learning_rate": 3.75713741910925e-06,
+      "loss": 1.3995,
+      "step": 961
+    },
+    {
+      "epoch": 0.9881869542886492,
+      "grad_norm": 2.367835760116577,
+      "learning_rate": 3.7552341073467836e-06,
+      "loss": 1.4491,
+      "step": 962
+    },
+    {
+      "epoch": 0.9892141756548536,
+      "grad_norm": 2.433840751647949,
+      "learning_rate": 3.753330795584317e-06,
+      "loss": 1.5078,
+      "step": 963
+    },
+    {
+      "epoch": 0.9902413970210581,
+      "grad_norm": 2.3976447582244873,
+      "learning_rate": 3.75142748382185e-06,
+      "loss": 1.4107,
+      "step": 964
+    },
+    {
+      "epoch": 0.9912686183872624,
+      "grad_norm": 2.4369053840637207,
+      "learning_rate": 3.749524172059384e-06,
+      "loss": 1.4569,
+      "step": 965
+    },
+    {
+      "epoch": 0.9922958397534669,
+      "grad_norm": 2.3203182220458984,
+      "learning_rate": 3.747620860296917e-06,
+      "loss": 1.4976,
+      "step": 966
+    },
+    {
+      "epoch": 0.9933230611196713,
+      "grad_norm": 2.431885242462158,
+      "learning_rate": 3.7457175485344504e-06,
+      "loss": 1.4945,
+      "step": 967
+    },
+    {
+      "epoch": 0.9943502824858758,
+      "grad_norm": 2.2997021675109863,
+      "learning_rate": 3.7438142367719837e-06,
+      "loss": 1.48,
+      "step": 968
+    },
+    {
+      "epoch": 0.9953775038520801,
+      "grad_norm": 2.2502849102020264,
+      "learning_rate": 3.741910925009517e-06,
+      "loss": 1.4322,
+      "step": 969
+    },
+    {
+      "epoch": 0.9964047252182845,
+      "grad_norm": 2.311049222946167,
+      "learning_rate": 3.7400076132470505e-06,
+      "loss": 1.4498,
+      "step": 970
+    },
+    {
+      "epoch": 0.9964047252182845,
+      "eval_loss": 1.771975040435791,
+      "eval_runtime": 37.852,
+      "eval_samples_per_second": 26.419,
+      "eval_steps_per_second": 1.11,
+      "step": 970
+    },
+    {
+      "epoch": 0.997431946584489,
+      "grad_norm": 2.431821584701538,
+      "learning_rate": 3.7381043014845834e-06,
+      "loss": 1.4796,
+      "step": 971
+    },
+    {
+      "epoch": 0.9984591679506933,
+      "grad_norm": 2.319145441055298,
+      "learning_rate": 3.7362009897221168e-06,
+      "loss": 1.459,
+      "step": 972
+    },
+    {
+      "epoch": 0.9994863893168978,
+      "grad_norm": 2.4445269107818604,
+      "learning_rate": 3.73429767795965e-06,
+      "loss": 1.4764,
+      "step": 973
+    },
+    {
+      "epoch": 1.0005136106831023,
+      "grad_norm": 2.272233009338379,
+      "learning_rate": 3.7323943661971835e-06,
+      "loss": 1.3799,
+      "step": 974
+    },
+    {
+      "epoch": 1.0015408320493067,
+      "grad_norm": 2.5421202182769775,
+      "learning_rate": 3.730491054434717e-06,
+      "loss": 1.3257,
+      "step": 975
+    },
+    {
+      "epoch": 1.002568053415511,
+      "grad_norm": 2.3988611698150635,
+      "learning_rate": 3.72858774267225e-06,
+      "loss": 1.4244,
+      "step": 976
+    },
+    {
+      "epoch": 1.0035952747817154,
+      "grad_norm": 2.5080084800720215,
+      "learning_rate": 3.726684430909783e-06,
+      "loss": 1.4461,
+      "step": 977
+    },
+    {
+      "epoch": 1.00462249614792,
+      "grad_norm": 2.411081314086914,
+      "learning_rate": 3.7247811191473165e-06,
+      "loss": 1.45,
+      "step": 978
+    },
+    {
+      "epoch": 1.0056497175141244,
+      "grad_norm": 2.729546070098877,
+      "learning_rate": 3.72287780738485e-06,
+      "loss": 1.4296,
+      "step": 979
+    },
+    {
+      "epoch": 1.0066769388803287,
+      "grad_norm": 2.5932579040527344,
+      "learning_rate": 3.7209744956223832e-06,
+      "loss": 1.4312,
+      "step": 980
+    },
+    {
+      "epoch": 1.0066769388803287,
+      "eval_loss": 1.770633339881897,
+      "eval_runtime": 37.8977,
+      "eval_samples_per_second": 26.387,
+      "eval_steps_per_second": 1.108,
+      "step": 980
+    },
+    {
+      "epoch": 1.007704160246533,
+      "grad_norm": 2.4512336254119873,
+      "learning_rate": 3.719071183859916e-06,
+      "loss": 1.3637,
+      "step": 981
+    },
+    {
+      "epoch": 1.0087313816127375,
+      "grad_norm": 2.5416886806488037,
+      "learning_rate": 3.7171678720974495e-06,
+      "loss": 1.4591,
+      "step": 982
+    },
+    {
+      "epoch": 1.009758602978942,
+      "grad_norm": 2.464679002761841,
+      "learning_rate": 3.715264560334983e-06,
+      "loss": 1.4306,
+      "step": 983
+    },
+    {
+      "epoch": 1.0107858243451464,
+      "grad_norm": 2.3745839595794678,
+      "learning_rate": 3.7133612485725163e-06,
+      "loss": 1.4426,
+      "step": 984
+    },
+    {
+      "epoch": 1.0118130457113508,
+      "grad_norm": 2.3606340885162354,
+      "learning_rate": 3.71145793681005e-06,
+      "loss": 1.3618,
+      "step": 985
+    },
+    {
+      "epoch": 1.0128402670775551,
+      "grad_norm": 2.5836591720581055,
+      "learning_rate": 3.7095546250475834e-06,
+      "loss": 1.4776,
+      "step": 986
+    },
+    {
+      "epoch": 1.0138674884437597,
+      "grad_norm": 2.4108684062957764,
+      "learning_rate": 3.7076513132851168e-06,
+      "loss": 1.4047,
+      "step": 987
+    },
+    {
+      "epoch": 1.014894709809964,
+      "grad_norm": 2.4316983222961426,
+      "learning_rate": 3.7057480015226497e-06,
+      "loss": 1.3772,
+      "step": 988
+    },
+    {
+      "epoch": 1.0159219311761685,
+      "grad_norm": 2.5510315895080566,
+      "learning_rate": 3.703844689760183e-06,
+      "loss": 1.4536,
+      "step": 989
+    },
+    {
+      "epoch": 1.0169491525423728,
+      "grad_norm": 2.4947116374969482,
+      "learning_rate": 3.7019413779977164e-06,
+      "loss": 1.434,
+      "step": 990
+    },
+    {
+      "epoch": 1.0169491525423728,
+      "eval_loss": 1.7928223609924316,
+      "eval_runtime": 37.8838,
+      "eval_samples_per_second": 26.397,
+      "eval_steps_per_second": 1.109,
+      "step": 990
+    },
+    {
+      "epoch": 1.0179763739085772,
+      "grad_norm": 2.5938122272491455,
+      "learning_rate": 3.70003806623525e-06,
+      "loss": 1.4083,
+      "step": 991
+    },
+    {
+      "epoch": 1.0190035952747818,
+      "grad_norm": 2.472865581512451,
+      "learning_rate": 3.698134754472783e-06,
+      "loss": 1.4696,
+      "step": 992
+    },
+    {
+      "epoch": 1.0200308166409862,
+      "grad_norm": 2.374344825744629,
+      "learning_rate": 3.696231442710316e-06,
+      "loss": 1.403,
+      "step": 993
+    },
+    {
+      "epoch": 1.0210580380071905,
+      "grad_norm": 2.6928529739379883,
+      "learning_rate": 3.6943281309478495e-06,
+      "loss": 1.4216,
+      "step": 994
+    },
+    {
+      "epoch": 1.022085259373395,
+      "grad_norm": 2.640890598297119,
+      "learning_rate": 3.692424819185383e-06,
+      "loss": 1.4362,
+      "step": 995
+    },
+    {
+      "epoch": 1.0231124807395995,
+      "grad_norm": 2.45357084274292,
+      "learning_rate": 3.690521507422916e-06,
+      "loss": 1.382,
+      "step": 996
+    },
+    {
+      "epoch": 1.0241397021058039,
+      "grad_norm": 2.712517023086548,
+      "learning_rate": 3.6886181956604496e-06,
+      "loss": 1.395,
+      "step": 997
+    },
+    {
+      "epoch": 1.0251669234720082,
+      "grad_norm": 2.7127771377563477,
+      "learning_rate": 3.686714883897983e-06,
+      "loss": 1.4589,
+      "step": 998
+    },
+    {
+      "epoch": 1.0261941448382126,
+      "grad_norm": 2.5379409790039062,
+      "learning_rate": 3.684811572135516e-06,
+      "loss": 1.4551,
+      "step": 999
+    },
+    {
+      "epoch": 1.027221366204417,
+      "grad_norm": 2.5015134811401367,
+      "learning_rate": 3.6829082603730492e-06,
+      "loss": 1.4112,
+      "step": 1000
+    },
+    {
+      "epoch": 1.027221366204417,
+      "eval_loss": 1.7806837558746338,
+      "eval_runtime": 37.9091,
+      "eval_samples_per_second": 26.379,
+      "eval_steps_per_second": 1.108,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0282485875706215,
+      "grad_norm": 2.481306314468384,
+      "learning_rate": 3.6810049486105826e-06,
+      "loss": 1.4469,
+      "step": 1001
+    },
+    {
+      "epoch": 1.029275808936826,
+      "grad_norm": 2.481520891189575,
+      "learning_rate": 3.679101636848116e-06,
+      "loss": 1.4284,
+      "step": 1002
+    },
+    {
+      "epoch": 1.0303030303030303,
+      "grad_norm": 2.4291019439697266,
+      "learning_rate": 3.6771983250856493e-06,
+      "loss": 1.4067,
+      "step": 1003
+    },
+    {
+      "epoch": 1.0313302516692346,
+      "grad_norm": 2.5361173152923584,
+      "learning_rate": 3.6752950133231822e-06,
+      "loss": 1.4347,
+      "step": 1004
+    },
+    {
+      "epoch": 1.0323574730354392,
+      "grad_norm": 2.5588085651397705,
+      "learning_rate": 3.6733917015607156e-06,
+      "loss": 1.4455,
+      "step": 1005
+    },
+    {
+      "epoch": 1.0333846944016436,
+      "grad_norm": 2.555091619491577,
+      "learning_rate": 3.6714883897982494e-06,
+      "loss": 1.399,
+      "step": 1006
+    },
+    {
+      "epoch": 1.034411915767848,
+      "grad_norm": 2.5489652156829834,
+      "learning_rate": 3.6695850780357828e-06,
+      "loss": 1.3894,
+      "step": 1007
+    },
+    {
+      "epoch": 1.0354391371340523,
+      "grad_norm": 2.5258426666259766,
+      "learning_rate": 3.667681766273316e-06,
+      "loss": 1.3962,
+      "step": 1008
+    },
+    {
+      "epoch": 1.0364663585002567,
+      "grad_norm": 2.455916166305542,
+      "learning_rate": 3.6657784545108495e-06,
+      "loss": 1.4773,
+      "step": 1009
+    },
+    {
+      "epoch": 1.0374935798664613,
+      "grad_norm": 2.5466208457946777,
+      "learning_rate": 3.6638751427483824e-06,
+      "loss": 1.4337,
+      "step": 1010
+    },
+    {
+      "epoch": 1.0374935798664613,
+      "eval_loss": 1.7840261459350586,
+      "eval_runtime": 37.6685,
+      "eval_samples_per_second": 26.547,
+      "eval_steps_per_second": 1.115,
+      "step": 1010
+    },
+    {
+      "epoch": 1.0385208012326657,
+      "grad_norm": 2.5446081161499023,
+      "learning_rate": 3.6619718309859158e-06,
+      "loss": 1.4777,
+      "step": 1011
+    },
+    {
+      "epoch": 1.03954802259887,
+      "grad_norm": 2.509493350982666,
+      "learning_rate": 3.660068519223449e-06,
+      "loss": 1.4295,
+      "step": 1012
+    },
+    {
+      "epoch": 1.0405752439650744,
+      "grad_norm": 2.417632818222046,
+      "learning_rate": 3.6581652074609825e-06,
+      "loss": 1.4044,
+      "step": 1013
+    },
+    {
+      "epoch": 1.041602465331279,
+      "grad_norm": 2.457902193069458,
+      "learning_rate": 3.656261895698516e-06,
+      "loss": 1.3948,
+      "step": 1014
+    },
+    {
+      "epoch": 1.0426296866974833,
+      "grad_norm": 2.4869253635406494,
+      "learning_rate": 3.6543585839360492e-06,
+      "loss": 1.397,
+      "step": 1015
+    },
+    {
+      "epoch": 1.0436569080636877,
+      "grad_norm": 2.495419979095459,
+      "learning_rate": 3.652455272173582e-06,
+      "loss": 1.4037,
+      "step": 1016
+    },
+    {
+      "epoch": 1.044684129429892,
+      "grad_norm": 2.5231642723083496,
+      "learning_rate": 3.6505519604111155e-06,
+      "loss": 1.4156,
+      "step": 1017
+    },
+    {
+      "epoch": 1.0457113507960965,
+      "grad_norm": 2.8302087783813477,
+      "learning_rate": 3.648648648648649e-06,
+      "loss": 1.4529,
+      "step": 1018
+    },
+    {
+      "epoch": 1.046738572162301,
+      "grad_norm": 2.467430353164673,
+      "learning_rate": 3.6467453368861823e-06,
+      "loss": 1.4078,
+      "step": 1019
+    },
+    {
+      "epoch": 1.0477657935285054,
+      "grad_norm": 2.469128131866455,
+      "learning_rate": 3.6448420251237156e-06,
+      "loss": 1.4344,
+      "step": 1020
+    },
+    {
+      "epoch": 1.0477657935285054,
+      "eval_loss": 1.7800015211105347,
+      "eval_runtime": 37.854,
+      "eval_samples_per_second": 26.417,
+      "eval_steps_per_second": 1.11,
+      "step": 1020
+    },
+    {
+      "epoch": 1.0487930148947098,
+      "grad_norm": 2.59389328956604,
+      "learning_rate": 3.6429387133612486e-06,
+      "loss": 1.451,
+      "step": 1021
+    },
+    {
+      "epoch": 1.0498202362609141,
+      "grad_norm": 2.5882468223571777,
+      "learning_rate": 3.641035401598782e-06,
+      "loss": 1.4026,
+      "step": 1022
+    },
+    {
+      "epoch": 1.0508474576271187,
+      "grad_norm": 2.601926565170288,
+      "learning_rate": 3.6391320898363153e-06,
+      "loss": 1.4382,
+      "step": 1023
+    },
+    {
+      "epoch": 1.051874678993323,
+      "grad_norm": 2.5336716175079346,
+      "learning_rate": 3.6372287780738487e-06,
+      "loss": 1.4184,
+      "step": 1024
+    },
+    {
+      "epoch": 1.0529019003595275,
+      "grad_norm": 2.503507614135742,
+      "learning_rate": 3.635325466311382e-06,
+      "loss": 1.4426,
+      "step": 1025
+    },
+    {
+      "epoch": 1.0539291217257318,
+      "grad_norm": 2.483013153076172,
+      "learning_rate": 3.633422154548916e-06,
+      "loss": 1.4563,
+      "step": 1026
+    },
+    {
+      "epoch": 1.0549563430919364,
+      "grad_norm": 2.4803354740142822,
+      "learning_rate": 3.6315188427864487e-06,
+      "loss": 1.4284,
+      "step": 1027
+    },
+    {
+      "epoch": 1.0559835644581408,
+      "grad_norm": 2.3735713958740234,
+      "learning_rate": 3.629615531023982e-06,
+      "loss": 1.3772,
+      "step": 1028
+    },
+    {
+      "epoch": 1.0570107858243452,
+      "grad_norm": 2.471226215362549,
+      "learning_rate": 3.6277122192615155e-06,
+      "loss": 1.4399,
+      "step": 1029
+    },
+    {
+      "epoch": 1.0580380071905495,
+      "grad_norm": 2.3827438354492188,
+      "learning_rate": 3.625808907499049e-06,
+      "loss": 1.3951,
+      "step": 1030
+    },
+    {
+      "epoch": 1.0580380071905495,
+      "eval_loss": 1.7787058353424072,
+      "eval_runtime": 37.9484,
+      "eval_samples_per_second": 26.352,
+      "eval_steps_per_second": 1.107,
+      "step": 1030
+    },
+    {
+      "epoch": 1.059065228556754,
+      "grad_norm": 2.403416633605957,
+      "learning_rate": 3.623905595736582e-06,
+      "loss": 1.3946,
+      "step": 1031
+    },
+    {
+      "epoch": 1.0600924499229585,
+      "grad_norm": 2.455740213394165,
+      "learning_rate": 3.6220022839741156e-06,
+      "loss": 1.3859,
+      "step": 1032
+    },
+    {
+      "epoch": 1.0611196712891628,
+      "grad_norm": 2.6349902153015137,
+      "learning_rate": 3.6200989722116485e-06,
+      "loss": 1.4092,
+      "step": 1033
+    },
+    {
+      "epoch": 1.0621468926553672,
+      "grad_norm": 2.6292974948883057,
+      "learning_rate": 3.618195660449182e-06,
+      "loss": 1.419,
+      "step": 1034
+    },
+    {
+      "epoch": 1.0631741140215716,
+      "grad_norm": 2.4338958263397217,
+      "learning_rate": 3.6162923486867152e-06,
+      "loss": 1.3546,
+      "step": 1035
+    },
+    {
+      "epoch": 1.064201335387776,
+      "grad_norm": 2.376420259475708,
+      "learning_rate": 3.6143890369242486e-06,
+      "loss": 1.3899,
+      "step": 1036
+    },
+    {
+      "epoch": 1.0652285567539805,
+      "grad_norm": 2.428738832473755,
+      "learning_rate": 3.612485725161782e-06,
+      "loss": 1.3955,
+      "step": 1037
+    },
+    {
+      "epoch": 1.066255778120185,
+      "grad_norm": 2.4316651821136475,
+      "learning_rate": 3.610582413399315e-06,
+      "loss": 1.3875,
+      "step": 1038
+    },
+    {
+      "epoch": 1.0672829994863893,
+      "grad_norm": 2.3969266414642334,
+      "learning_rate": 3.6086791016368482e-06,
+      "loss": 1.404,
+      "step": 1039
+    },
+    {
+      "epoch": 1.0683102208525936,
+      "grad_norm": 2.662283420562744,
+      "learning_rate": 3.6067757898743816e-06,
+      "loss": 1.4639,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0683102208525936,
+      "eval_loss": 1.7603929042816162,
+      "eval_runtime": 38.0098,
+      "eval_samples_per_second": 26.309,
+      "eval_steps_per_second": 1.105,
+      "step": 1040
+    },
+    {
+      "epoch": 1.0693374422187982,
+      "grad_norm": 2.601452112197876,
+      "learning_rate": 3.604872478111915e-06,
+      "loss": 1.47,
+      "step": 1041
+    },
+    {
+      "epoch": 1.0703646635850026,
+      "grad_norm": 2.4221863746643066,
+      "learning_rate": 3.6029691663494483e-06,
+      "loss": 1.4162,
+      "step": 1042
+    },
+    {
+      "epoch": 1.071391884951207,
+      "grad_norm": 2.4996650218963623,
+      "learning_rate": 3.6010658545869813e-06,
+      "loss": 1.443,
+      "step": 1043
+    },
+    {
+      "epoch": 1.0724191063174113,
+      "grad_norm": 2.547239303588867,
+      "learning_rate": 3.5991625428245146e-06,
+      "loss": 1.4464,
+      "step": 1044
+    },
+    {
+      "epoch": 1.073446327683616,
+      "grad_norm": 2.481187343597412,
+      "learning_rate": 3.597259231062048e-06,
+      "loss": 1.3749,
+      "step": 1045
+    },
+    {
+      "epoch": 1.0744735490498203,
+      "grad_norm": 2.5504934787750244,
+      "learning_rate": 3.5953559192995818e-06,
+      "loss": 1.4231,
+      "step": 1046
+    },
+    {
+      "epoch": 1.0755007704160247,
+      "grad_norm": 2.465390205383301,
+      "learning_rate": 3.593452607537115e-06,
+      "loss": 1.4805,
+      "step": 1047
+    },
+    {
+      "epoch": 1.076527991782229,
+      "grad_norm": 2.4399969577789307,
+      "learning_rate": 3.5915492957746485e-06,
+      "loss": 1.4091,
+      "step": 1048
+    },
+    {
+      "epoch": 1.0775552131484334,
+      "grad_norm": 2.4819061756134033,
+      "learning_rate": 3.589645984012182e-06,
+      "loss": 1.4367,
+      "step": 1049
+    },
+    {
+      "epoch": 1.078582434514638,
+      "grad_norm": 2.459517002105713,
+      "learning_rate": 3.587742672249715e-06,
+      "loss": 1.3882,
+      "step": 1050
+    },
+    {
+      "epoch": 1.078582434514638,
+      "eval_loss": 1.7557458877563477,
+      "eval_runtime": 38.0271,
+      "eval_samples_per_second": 26.297,
+      "eval_steps_per_second": 1.104,
+      "step": 1050
+    },
+    {
+      "epoch": 1.0796096558808423,
+      "grad_norm": 2.5434813499450684,
+      "learning_rate": 3.585839360487248e-06,
+      "loss": 1.3926,
+      "step": 1051
+    },
+    {
+      "epoch": 1.0806368772470467,
+      "grad_norm": 2.6173205375671387,
+      "learning_rate": 3.5839360487247815e-06,
+      "loss": 1.4044,
+      "step": 1052
+    },
+    {
+      "epoch": 1.081664098613251,
+      "grad_norm": 2.677584171295166,
+      "learning_rate": 3.582032736962315e-06,
+      "loss": 1.4695,
+      "step": 1053
+    },
+    {
+      "epoch": 1.0826913199794557,
+      "grad_norm": 2.4151623249053955,
+      "learning_rate": 3.5801294251998483e-06,
+      "loss": 1.3699,
+      "step": 1054
+    },
+    {
+      "epoch": 1.08371854134566,
+      "grad_norm": 2.4683942794799805,
+      "learning_rate": 3.578226113437381e-06,
+      "loss": 1.4132,
+      "step": 1055
+    },
+    {
+      "epoch": 1.0847457627118644,
+      "grad_norm": 2.6076581478118896,
+      "learning_rate": 3.5763228016749146e-06,
+      "loss": 1.4729,
+      "step": 1056
+    },
+    {
+      "epoch": 1.0857729840780688,
+      "grad_norm": 2.4766876697540283,
+      "learning_rate": 3.574419489912448e-06,
+      "loss": 1.4196,
+      "step": 1057
+    },
+    {
+      "epoch": 1.0868002054442731,
+      "grad_norm": 2.466458559036255,
+      "learning_rate": 3.5725161781499813e-06,
+      "loss": 1.4049,
+      "step": 1058
+    },
+    {
+      "epoch": 1.0878274268104777,
+      "grad_norm": 2.4220409393310547,
+      "learning_rate": 3.5706128663875147e-06,
+      "loss": 1.4246,
+      "step": 1059
+    },
+    {
+      "epoch": 1.088854648176682,
+      "grad_norm": 2.5571770668029785,
+      "learning_rate": 3.5687095546250476e-06,
+      "loss": 1.4397,
+      "step": 1060
+    },
+    {
+      "epoch": 1.088854648176682,
+      "eval_loss": 1.7701127529144287,
+      "eval_runtime": 38.0754,
+      "eval_samples_per_second": 26.264,
+      "eval_steps_per_second": 1.103,
+      "step": 1060
+    },
+    {
+      "epoch": 1.0898818695428865,
+      "grad_norm": 2.648768186569214,
+      "learning_rate": 3.566806242862581e-06,
+      "loss": 1.423,
+      "step": 1061
+    },
+    {
+      "epoch": 1.0909090909090908,
+      "grad_norm": 2.403651714324951,
+      "learning_rate": 3.5649029311001143e-06,
+      "loss": 1.3995,
+      "step": 1062
+    },
+    {
+      "epoch": 1.0919363122752954,
+      "grad_norm": 2.465538263320923,
+      "learning_rate": 3.5629996193376477e-06,
+      "loss": 1.4139,
+      "step": 1063
+    },
+    {
+      "epoch": 1.0929635336414998,
+      "grad_norm": 2.464662551879883,
+      "learning_rate": 3.561096307575181e-06,
+      "loss": 1.4499,
+      "step": 1064
+    },
+    {
+      "epoch": 1.0939907550077042,
+      "grad_norm": 2.4943597316741943,
+      "learning_rate": 3.559192995812714e-06,
+      "loss": 1.4555,
+      "step": 1065
+    },
+    {
+      "epoch": 1.0950179763739085,
+      "grad_norm": 2.6332199573516846,
+      "learning_rate": 3.557289684050248e-06,
+      "loss": 1.4484,
+      "step": 1066
+    },
+    {
+      "epoch": 1.0960451977401129,
+      "grad_norm": 2.4370667934417725,
+      "learning_rate": 3.555386372287781e-06,
+      "loss": 1.4588,
+      "step": 1067
+    },
+    {
+      "epoch": 1.0970724191063175,
+      "grad_norm": 2.568368673324585,
+      "learning_rate": 3.5534830605253145e-06,
+      "loss": 1.4269,
+      "step": 1068
+    },
+    {
+      "epoch": 1.0980996404725218,
+      "grad_norm": 2.4618077278137207,
+      "learning_rate": 3.551579748762848e-06,
+      "loss": 1.4165,
+      "step": 1069
+    },
+    {
+      "epoch": 1.0991268618387262,
+      "grad_norm": 2.551320791244507,
+      "learning_rate": 3.5496764370003812e-06,
+      "loss": 1.3991,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0991268618387262,
+      "eval_loss": 1.7648736238479614,
+      "eval_runtime": 38.0273,
+      "eval_samples_per_second": 26.297,
+      "eval_steps_per_second": 1.104,
+      "step": 1070
+    },
+    {
+      "epoch": 1.1001540832049306,
+      "grad_norm": 2.350905418395996,
+      "learning_rate": 3.5477731252379146e-06,
+      "loss": 1.3556,
+      "step": 1071
+    },
+    {
+      "epoch": 1.1011813045711352,
+      "grad_norm": 2.4294588565826416,
+      "learning_rate": 3.5458698134754475e-06,
+      "loss": 1.3801,
+      "step": 1072
+    },
+    {
+      "epoch": 1.1022085259373395,
+      "grad_norm": 2.576774835586548,
+      "learning_rate": 3.543966501712981e-06,
+      "loss": 1.3876,
+      "step": 1073
+    },
+    {
+      "epoch": 1.103235747303544,
+      "grad_norm": 2.716935873031616,
+      "learning_rate": 3.5420631899505142e-06,
+      "loss": 1.4907,
+      "step": 1074
+    },
+    {
+      "epoch": 1.1042629686697483,
+      "grad_norm": 2.425305128097534,
+      "learning_rate": 3.5401598781880476e-06,
+      "loss": 1.3161,
+      "step": 1075
+    },
+    {
+      "epoch": 1.1052901900359529,
+      "grad_norm": 2.54738712310791,
+      "learning_rate": 3.538256566425581e-06,
+      "loss": 1.4462,
+      "step": 1076
+    },
+    {
+      "epoch": 1.1063174114021572,
+      "grad_norm": 2.5109341144561768,
+      "learning_rate": 3.536353254663114e-06,
+      "loss": 1.4022,
+      "step": 1077
+    },
+    {
+      "epoch": 1.1073446327683616,
+      "grad_norm": 2.468749523162842,
+      "learning_rate": 3.5344499429006473e-06,
+      "loss": 1.3857,
+      "step": 1078
+    },
+    {
+      "epoch": 1.108371854134566,
+      "grad_norm": 2.5285134315490723,
+      "learning_rate": 3.5325466311381806e-06,
+      "loss": 1.4309,
+      "step": 1079
+    },
+    {
+      "epoch": 1.1093990755007703,
+      "grad_norm": 2.6391055583953857,
+      "learning_rate": 3.530643319375714e-06,
+      "loss": 1.3644,
+      "step": 1080
+    },
+    {
+      "epoch": 1.1093990755007703,
+      "eval_loss": 1.7846598625183105,
+      "eval_runtime": 38.0604,
+      "eval_samples_per_second": 26.274,
+      "eval_steps_per_second": 1.104,
+      "step": 1080
+    },
+    {
+      "epoch": 1.110426296866975,
+      "grad_norm": 2.660356283187866,
+      "learning_rate": 3.5287400076132474e-06,
+      "loss": 1.3922,
+      "step": 1081
+    },
+    {
+      "epoch": 1.1114535182331793,
+      "grad_norm": 2.5648741722106934,
+      "learning_rate": 3.5268366958507803e-06,
+      "loss": 1.4282,
+      "step": 1082
+    },
+    {
+      "epoch": 1.1124807395993837,
+      "grad_norm": 2.6862876415252686,
+      "learning_rate": 3.5249333840883137e-06,
+      "loss": 1.4208,
+      "step": 1083
+    },
+    {
+      "epoch": 1.113507960965588,
+      "grad_norm": 2.6200191974639893,
+      "learning_rate": 3.523030072325847e-06,
+      "loss": 1.3883,
+      "step": 1084
+    },
+    {
+      "epoch": 1.1145351823317924,
+      "grad_norm": 2.5802111625671387,
+      "learning_rate": 3.5211267605633804e-06,
+      "loss": 1.4033,
+      "step": 1085
+    },
+    {
+      "epoch": 1.115562403697997,
+      "grad_norm": 2.4148080348968506,
+      "learning_rate": 3.519223448800914e-06,
+      "loss": 1.4085,
+      "step": 1086
+    },
+    {
+      "epoch": 1.1165896250642013,
+      "grad_norm": 2.7452127933502197,
+      "learning_rate": 3.5173201370384475e-06,
+      "loss": 1.4021,
+      "step": 1087
+    },
+    {
+      "epoch": 1.1176168464304057,
+      "grad_norm": 2.6437106132507324,
+      "learning_rate": 3.515416825275981e-06,
+      "loss": 1.4161,
+      "step": 1088
+    },
+    {
+      "epoch": 1.11864406779661,
+      "grad_norm": 2.4709842205047607,
+      "learning_rate": 3.513513513513514e-06,
+      "loss": 1.4322,
+      "step": 1089
+    },
+    {
+      "epoch": 1.1196712891628147,
+      "grad_norm": 2.5138099193573,
+      "learning_rate": 3.511610201751047e-06,
+      "loss": 1.3435,
+      "step": 1090
+    },
+    {
+      "epoch": 1.1196712891628147,
+      "eval_loss": 1.7687023878097534,
+      "eval_runtime": 38.0137,
+      "eval_samples_per_second": 26.306,
+      "eval_steps_per_second": 1.105,
+      "step": 1090
+    },
+    {
+      "epoch": 1.120698510529019,
+      "grad_norm": 2.4194881916046143,
+      "learning_rate": 3.5097068899885806e-06,
+      "loss": 1.4133,
+      "step": 1091
+    },
+    {
+      "epoch": 1.1217257318952234,
+      "grad_norm": 2.6564371585845947,
+      "learning_rate": 3.507803578226114e-06,
+      "loss": 1.4759,
+      "step": 1092
+    },
+    {
+      "epoch": 1.1227529532614278,
+      "grad_norm": 2.4480180740356445,
+      "learning_rate": 3.5059002664636473e-06,
+      "loss": 1.3577,
+      "step": 1093
+    },
+    {
+      "epoch": 1.1237801746276324,
+      "grad_norm": 2.5538978576660156,
+      "learning_rate": 3.5039969547011802e-06,
+      "loss": 1.3984,
+      "step": 1094
+    },
+    {
+      "epoch": 1.1248073959938367,
+      "grad_norm": 2.6467089653015137,
+      "learning_rate": 3.5020936429387136e-06,
+      "loss": 1.4049,
+      "step": 1095
+    },
+    {
+      "epoch": 1.125834617360041,
+      "grad_norm": 2.6502673625946045,
+      "learning_rate": 3.500190331176247e-06,
+      "loss": 1.462,
+      "step": 1096
+    },
+    {
+      "epoch": 1.1268618387262455,
+      "grad_norm": 2.5696403980255127,
+      "learning_rate": 3.4982870194137803e-06,
+      "loss": 1.3926,
+      "step": 1097
+    },
+    {
+      "epoch": 1.1278890600924498,
+      "grad_norm": 2.5951733589172363,
+      "learning_rate": 3.4963837076513137e-06,
+      "loss": 1.4031,
+      "step": 1098
+    },
+    {
+      "epoch": 1.1289162814586544,
+      "grad_norm": 2.4910218715667725,
+      "learning_rate": 3.4944803958888466e-06,
+      "loss": 1.3348,
+      "step": 1099
+    },
+    {
+      "epoch": 1.1299435028248588,
+      "grad_norm": 2.6325888633728027,
+      "learning_rate": 3.49257708412638e-06,
+      "loss": 1.4641,
+      "step": 1100
+    },
+    {
+      "epoch": 1.1299435028248588,
+      "eval_loss": 1.7720683813095093,
+      "eval_runtime": 37.9585,
+      "eval_samples_per_second": 26.345,
+      "eval_steps_per_second": 1.106,
+      "step": 1100
+    },
+    {
+      "epoch": 1.1309707241910631,
+      "grad_norm": 2.6712775230407715,
+      "learning_rate": 3.4906737723639133e-06,
+      "loss": 1.4726,
+      "step": 1101
+    },
+    {
+      "epoch": 1.1319979455572675,
+      "grad_norm": 2.580404043197632,
+      "learning_rate": 3.4887704606014467e-06,
+      "loss": 1.4158,
+      "step": 1102
+    },
+    {
+      "epoch": 1.133025166923472,
+      "grad_norm": 2.467453718185425,
+      "learning_rate": 3.48686714883898e-06,
+      "loss": 1.4636,
+      "step": 1103
+    },
+    {
+      "epoch": 1.1340523882896765,
+      "grad_norm": 2.627136468887329,
+      "learning_rate": 3.484963837076513e-06,
+      "loss": 1.4082,
+      "step": 1104
+    },
+    {
+      "epoch": 1.1350796096558808,
+      "grad_norm": 2.6198604106903076,
+      "learning_rate": 3.4830605253140464e-06,
+      "loss": 1.3628,
+      "step": 1105
+    },
+    {
+      "epoch": 1.1361068310220852,
+      "grad_norm": 2.439512014389038,
+      "learning_rate": 3.4811572135515797e-06,
+      "loss": 1.3885,
+      "step": 1106
+    },
+    {
+      "epoch": 1.1371340523882898,
+      "grad_norm": 2.5543010234832764,
+      "learning_rate": 3.4792539017891135e-06,
+      "loss": 1.3674,
+      "step": 1107
+    },
+    {
+      "epoch": 1.1381612737544942,
+      "grad_norm": 2.7263364791870117,
+      "learning_rate": 3.477350590026647e-06,
+      "loss": 1.4387,
+      "step": 1108
+    },
+    {
+      "epoch": 1.1391884951206985,
+      "grad_norm": 2.5030503273010254,
+      "learning_rate": 3.4754472782641802e-06,
+      "loss": 1.4222,
+      "step": 1109
+    },
+    {
+      "epoch": 1.140215716486903,
+      "grad_norm": 2.5090320110321045,
+      "learning_rate": 3.4735439665017136e-06,
+      "loss": 1.3981,
+      "step": 1110
+    },
+    {
+      "epoch": 1.140215716486903,
+      "eval_loss": 1.769408941268921,
+      "eval_runtime": 37.608,
+      "eval_samples_per_second": 26.59,
+      "eval_steps_per_second": 1.117,
+      "step": 1110
+    },
+    {
+      "epoch": 1.1412429378531073,
+      "grad_norm": 2.575934648513794,
+      "learning_rate": 3.4716406547392465e-06,
+      "loss": 1.4074,
+      "step": 1111
+    },
+    {
+      "epoch": 1.1422701592193119,
+      "grad_norm": 2.6381359100341797,
+      "learning_rate": 3.46973734297678e-06,
+      "loss": 1.4162,
+      "step": 1112
+    },
+    {
+      "epoch": 1.1432973805855162,
+      "grad_norm": 2.7605674266815186,
+      "learning_rate": 3.4678340312143133e-06,
+      "loss": 1.4616,
+      "step": 1113
+    },
+    {
+      "epoch": 1.1443246019517206,
+      "grad_norm": 2.5548479557037354,
+      "learning_rate": 3.4659307194518466e-06,
+      "loss": 1.4511,
+      "step": 1114
+    },
+    {
+      "epoch": 1.145351823317925,
+      "grad_norm": 2.471168279647827,
+      "learning_rate": 3.46402740768938e-06,
+      "loss": 1.3777,
+      "step": 1115
+    },
+    {
+      "epoch": 1.1463790446841293,
+      "grad_norm": 2.4416346549987793,
+      "learning_rate": 3.462124095926913e-06,
+      "loss": 1.3805,
+      "step": 1116
+    },
+    {
+      "epoch": 1.147406266050334,
+      "grad_norm": 2.4972171783447266,
+      "learning_rate": 3.4602207841644463e-06,
+      "loss": 1.4282,
+      "step": 1117
+    },
+    {
+      "epoch": 1.1484334874165383,
+      "grad_norm": 2.4379162788391113,
+      "learning_rate": 3.4583174724019797e-06,
+      "loss": 1.4744,
+      "step": 1118
+    },
+    {
+      "epoch": 1.1494607087827426,
+      "grad_norm": 2.508047580718994,
+      "learning_rate": 3.456414160639513e-06,
+      "loss": 1.4303,
+      "step": 1119
+    },
+    {
+      "epoch": 1.150487930148947,
+      "grad_norm": 2.4299631118774414,
+      "learning_rate": 3.4545108488770464e-06,
+      "loss": 1.4179,
+      "step": 1120
+    },
+    {
+      "epoch": 1.150487930148947,
+      "eval_loss": 1.7768158912658691,
+      "eval_runtime": 37.736,
+      "eval_samples_per_second": 26.5,
+      "eval_steps_per_second": 1.113,
+      "step": 1120
+    },
+    {
+      "epoch": 1.1515151515151516,
+      "grad_norm": 2.41538143157959,
+      "learning_rate": 3.4526075371145793e-06,
+      "loss": 1.4235,
+      "step": 1121
+    },
+    {
+      "epoch": 1.152542372881356,
+      "grad_norm": 2.5614655017852783,
+      "learning_rate": 3.4507042253521127e-06,
+      "loss": 1.4879,
+      "step": 1122
+    },
+    {
+      "epoch": 1.1535695942475603,
+      "grad_norm": 2.4826605319976807,
+      "learning_rate": 3.448800913589646e-06,
+      "loss": 1.3968,
+      "step": 1123
+    },
+    {
+      "epoch": 1.1545968156137647,
+      "grad_norm": 2.5341391563415527,
+      "learning_rate": 3.4468976018271794e-06,
+      "loss": 1.4903,
+      "step": 1124
+    },
+    {
+      "epoch": 1.1556240369799693,
+      "grad_norm": 2.493818998336792,
+      "learning_rate": 3.4449942900647128e-06,
+      "loss": 1.4466,
+      "step": 1125
+    },
+    {
+      "epoch": 1.1566512583461737,
+      "grad_norm": 2.6057965755462646,
+      "learning_rate": 3.443090978302246e-06,
+      "loss": 1.4425,
+      "step": 1126
+    },
+    {
+      "epoch": 1.157678479712378,
+      "grad_norm": 2.5481390953063965,
+      "learning_rate": 3.44118766653978e-06,
+      "loss": 1.4296,
+      "step": 1127
+    },
+    {
+      "epoch": 1.1587057010785824,
+      "grad_norm": 2.5753650665283203,
+      "learning_rate": 3.439284354777313e-06,
+      "loss": 1.4408,
+      "step": 1128
+    },
+    {
+      "epoch": 1.1597329224447868,
+      "grad_norm": 2.5423803329467773,
+      "learning_rate": 3.4373810430148462e-06,
+      "loss": 1.373,
+      "step": 1129
+    },
+    {
+      "epoch": 1.1607601438109914,
+      "grad_norm": 2.593487024307251,
+      "learning_rate": 3.4354777312523796e-06,
+      "loss": 1.3992,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1607601438109914,
+      "eval_loss": 1.7784411907196045,
+      "eval_runtime": 37.8687,
+      "eval_samples_per_second": 26.407,
+      "eval_steps_per_second": 1.109,
+      "step": 1130
+    },
+    {
+      "epoch": 1.1617873651771957,
+      "grad_norm": 2.4693503379821777,
+      "learning_rate": 3.433574419489913e-06,
+      "loss": 1.4339,
+      "step": 1131
+    },
+    {
+      "epoch": 1.1628145865434,
+      "grad_norm": 2.5477423667907715,
+      "learning_rate": 3.4316711077274463e-06,
+      "loss": 1.4035,
+      "step": 1132
+    },
+    {
+      "epoch": 1.1638418079096045,
+      "grad_norm": 2.5650672912597656,
+      "learning_rate": 3.4297677959649792e-06,
+      "loss": 1.4517,
+      "step": 1133
+    },
+    {
+      "epoch": 1.1648690292758088,
+      "grad_norm": 2.3217973709106445,
+      "learning_rate": 3.4278644842025126e-06,
+      "loss": 1.3983,
+      "step": 1134
+    },
+    {
+      "epoch": 1.1658962506420134,
+      "grad_norm": 2.674342155456543,
+      "learning_rate": 3.425961172440046e-06,
+      "loss": 1.4799,
+      "step": 1135
+    },
+    {
+      "epoch": 1.1669234720082178,
+      "grad_norm": 2.4861397743225098,
+      "learning_rate": 3.4240578606775793e-06,
+      "loss": 1.414,
+      "step": 1136
+    },
+    {
+      "epoch": 1.1679506933744221,
+      "grad_norm": 2.4306106567382812,
+      "learning_rate": 3.4221545489151127e-06,
+      "loss": 1.3918,
+      "step": 1137
+    },
+    {
+      "epoch": 1.1689779147406265,
+      "grad_norm": 2.6873955726623535,
+      "learning_rate": 3.4202512371526456e-06,
+      "loss": 1.4979,
+      "step": 1138
+    },
+    {
+      "epoch": 1.170005136106831,
+      "grad_norm": 2.559042453765869,
+      "learning_rate": 3.418347925390179e-06,
+      "loss": 1.4424,
+      "step": 1139
+    },
+    {
+      "epoch": 1.1710323574730355,
+      "grad_norm": 2.397489547729492,
+      "learning_rate": 3.4164446136277124e-06,
+      "loss": 1.418,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1710323574730355,
+      "eval_loss": 1.7702736854553223,
+      "eval_runtime": 37.8963,
+      "eval_samples_per_second": 26.388,
+      "eval_steps_per_second": 1.108,
+      "step": 1140
+    },
+    {
+      "epoch": 1.1720595788392398,
+      "grad_norm": 2.4295787811279297,
+      "learning_rate": 3.4145413018652457e-06,
+      "loss": 1.4436,
+      "step": 1141
+    },
+    {
+      "epoch": 1.1730868002054442,
+      "grad_norm": 2.503202199935913,
+      "learning_rate": 3.412637990102779e-06,
+      "loss": 1.4598,
+      "step": 1142
+    },
+    {
+      "epoch": 1.1741140215716488,
+      "grad_norm": 2.540195941925049,
+      "learning_rate": 3.4107346783403125e-06,
+      "loss": 1.3704,
+      "step": 1143
+    },
+    {
+      "epoch": 1.1751412429378532,
+      "grad_norm": 2.4832522869110107,
+      "learning_rate": 3.4088313665778454e-06,
+      "loss": 1.4081,
+      "step": 1144
+    },
+    {
+      "epoch": 1.1761684643040575,
+      "grad_norm": 2.5639121532440186,
+      "learning_rate": 3.4069280548153788e-06,
+      "loss": 1.3866,
+      "step": 1145
+    },
+    {
+      "epoch": 1.177195685670262,
+      "grad_norm": 2.526610851287842,
+      "learning_rate": 3.405024743052912e-06,
+      "loss": 1.4181,
+      "step": 1146
+    },
+    {
+      "epoch": 1.1782229070364663,
+      "grad_norm": 2.6945254802703857,
+      "learning_rate": 3.403121431290446e-06,
+      "loss": 1.3871,
+      "step": 1147
+    },
+    {
+      "epoch": 1.1792501284026708,
+      "grad_norm": 2.6561484336853027,
+      "learning_rate": 3.4012181195279793e-06,
+      "loss": 1.3987,
+      "step": 1148
+    },
+    {
+      "epoch": 1.1802773497688752,
+      "grad_norm": 2.6104650497436523,
+      "learning_rate": 3.3993148077655126e-06,
+      "loss": 1.4057,
+      "step": 1149
+    },
+    {
+      "epoch": 1.1813045711350796,
+      "grad_norm": 2.571707248687744,
+      "learning_rate": 3.3974114960030456e-06,
+      "loss": 1.3606,
+      "step": 1150
+    },
+    {
+      "epoch": 1.1813045711350796,
+      "eval_loss": 1.7780840396881104,
+      "eval_runtime": 37.8616,
+      "eval_samples_per_second": 26.412,
+      "eval_steps_per_second": 1.109,
+      "step": 1150
+    },
+    {
+      "epoch": 1.182331792501284,
+      "grad_norm": 2.5804402828216553,
+      "learning_rate": 3.395508184240579e-06,
+      "loss": 1.3876,
+      "step": 1151
+    },
+    {
+      "epoch": 1.1833590138674885,
+      "grad_norm": 2.832446336746216,
+      "learning_rate": 3.3936048724781123e-06,
+      "loss": 1.4674,
+      "step": 1152
+    },
+    {
+      "epoch": 1.184386235233693,
+      "grad_norm": 2.6478688716888428,
+      "learning_rate": 3.3917015607156457e-06,
+      "loss": 1.3643,
+      "step": 1153
+    },
+    {
+      "epoch": 1.1854134565998973,
+      "grad_norm": 2.579416513442993,
+      "learning_rate": 3.389798248953179e-06,
+      "loss": 1.3874,
+      "step": 1154
+    },
+    {
+      "epoch": 1.1864406779661016,
+      "grad_norm": 2.487861156463623,
+      "learning_rate": 3.387894937190712e-06,
+      "loss": 1.4323,
+      "step": 1155
+    },
+    {
+      "epoch": 1.1874678993323062,
+      "grad_norm": 2.5020716190338135,
+      "learning_rate": 3.3859916254282453e-06,
+      "loss": 1.423,
+      "step": 1156
+    },
+    {
+      "epoch": 1.1884951206985106,
+      "grad_norm": 2.6588852405548096,
+      "learning_rate": 3.3840883136657787e-06,
+      "loss": 1.4808,
+      "step": 1157
+    },
+    {
+      "epoch": 1.189522342064715,
+      "grad_norm": 2.5702178478240967,
+      "learning_rate": 3.382185001903312e-06,
+      "loss": 1.4007,
+      "step": 1158
+    },
+    {
+      "epoch": 1.1905495634309193,
+      "grad_norm": 2.5082061290740967,
+      "learning_rate": 3.3802816901408454e-06,
+      "loss": 1.4278,
+      "step": 1159
+    },
+    {
+      "epoch": 1.1915767847971237,
+      "grad_norm": 2.4165053367614746,
+      "learning_rate": 3.3783783783783788e-06,
+      "loss": 1.4088,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1915767847971237,
+      "eval_loss": 1.7722209692001343,
+      "eval_runtime": 37.8843,
+      "eval_samples_per_second": 26.396,
+      "eval_steps_per_second": 1.109,
+      "step": 1160
+    },
+    {
+      "epoch": 1.1926040061633283,
+      "grad_norm": 2.4510512351989746,
+      "learning_rate": 3.3764750666159117e-06,
+      "loss": 1.4089,
+      "step": 1161
+    },
+    {
+      "epoch": 1.1936312275295327,
+      "grad_norm": 2.5006556510925293,
+      "learning_rate": 3.374571754853445e-06,
+      "loss": 1.3988,
+      "step": 1162
+    },
+    {
+      "epoch": 1.194658448895737,
+      "grad_norm": 2.416264057159424,
+      "learning_rate": 3.3726684430909784e-06,
+      "loss": 1.3576,
+      "step": 1163
+    },
+    {
+      "epoch": 1.1956856702619414,
+      "grad_norm": 2.465613842010498,
+      "learning_rate": 3.370765131328512e-06,
+      "loss": 1.3796,
+      "step": 1164
+    },
+    {
+      "epoch": 1.1967128916281458,
+      "grad_norm": 2.510542154312134,
+      "learning_rate": 3.368861819566045e-06,
+      "loss": 1.391,
+      "step": 1165
+    },
+    {
+      "epoch": 1.1977401129943503,
+      "grad_norm": 2.729495048522949,
+      "learning_rate": 3.366958507803578e-06,
+      "loss": 1.431,
+      "step": 1166
+    },
+    {
+      "epoch": 1.1987673343605547,
+      "grad_norm": 2.4473495483398438,
+      "learning_rate": 3.365055196041112e-06,
+      "loss": 1.376,
+      "step": 1167
+    },
+    {
+      "epoch": 1.199794555726759,
+      "grad_norm": 2.3869211673736572,
+      "learning_rate": 3.3631518842786452e-06,
+      "loss": 1.4498,
+      "step": 1168
+    },
+    {
+      "epoch": 1.2008217770929634,
+      "grad_norm": 2.4202256202697754,
+      "learning_rate": 3.3612485725161786e-06,
+      "loss": 1.3754,
+      "step": 1169
+    },
+    {
+      "epoch": 1.201848998459168,
+      "grad_norm": 2.452707529067993,
+      "learning_rate": 3.359345260753712e-06,
+      "loss": 1.4647,
+      "step": 1170
+    },
+    {
+      "epoch": 1.201848998459168,
+      "eval_loss": 1.769252896308899,
+      "eval_runtime": 37.9385,
+      "eval_samples_per_second": 26.358,
+      "eval_steps_per_second": 1.107,
+      "step": 1170
+    },
+    {
+      "epoch": 1.2028762198253724,
+      "grad_norm": 2.56616473197937,
+      "learning_rate": 3.3574419489912453e-06,
+      "loss": 1.443,
+      "step": 1171
+    },
+    {
+      "epoch": 1.2039034411915768,
+      "grad_norm": 2.6483540534973145,
+      "learning_rate": 3.3555386372287783e-06,
+      "loss": 1.4755,
+      "step": 1172
+    },
+    {
+      "epoch": 1.2049306625577811,
+      "grad_norm": 2.5579090118408203,
+      "learning_rate": 3.3536353254663116e-06,
+      "loss": 1.3732,
+      "step": 1173
+    },
+    {
+      "epoch": 1.2059578839239857,
+      "grad_norm": 2.5310683250427246,
+      "learning_rate": 3.351732013703845e-06,
+      "loss": 1.4077,
+      "step": 1174
+    },
+    {
+      "epoch": 1.20698510529019,
+      "grad_norm": 2.616957187652588,
+      "learning_rate": 3.3498287019413784e-06,
+      "loss": 1.4237,
+      "step": 1175
+    },
+    {
+      "epoch": 1.2080123266563945,
+      "grad_norm": 2.6008999347686768,
+      "learning_rate": 3.3479253901789117e-06,
+      "loss": 1.4381,
+      "step": 1176
+    },
+    {
+      "epoch": 1.2090395480225988,
+      "grad_norm": 2.5235824584960938,
+      "learning_rate": 3.346022078416445e-06,
+      "loss": 1.3471,
+      "step": 1177
+    },
+    {
+      "epoch": 1.2100667693888032,
+      "grad_norm": 2.3749704360961914,
+      "learning_rate": 3.344118766653978e-06,
+      "loss": 1.417,
+      "step": 1178
+    },
+    {
+      "epoch": 1.2110939907550078,
+      "grad_norm": 2.493380069732666,
+      "learning_rate": 3.3422154548915114e-06,
+      "loss": 1.3924,
+      "step": 1179
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "grad_norm": 2.4233086109161377,
+      "learning_rate": 3.3403121431290448e-06,
+      "loss": 1.4257,
+      "step": 1180
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "eval_loss": 1.7660146951675415,
+      "eval_runtime": 37.9268,
+      "eval_samples_per_second": 26.367,
+      "eval_steps_per_second": 1.107,
+      "step": 1180
+    },
+    {
+      "epoch": 1.2131484334874165,
+      "grad_norm": 2.535982608795166,
+      "learning_rate": 3.338408831366578e-06,
+      "loss": 1.3587,
+      "step": 1181
+    },
+    {
+      "epoch": 1.214175654853621,
+      "grad_norm": 2.5744845867156982,
+      "learning_rate": 3.3365055196041115e-06,
+      "loss": 1.436,
+      "step": 1182
+    },
+    {
+      "epoch": 1.2152028762198253,
+      "grad_norm": 2.5544042587280273,
+      "learning_rate": 3.3346022078416444e-06,
+      "loss": 1.464,
+      "step": 1183
+    },
+    {
+      "epoch": 1.2162300975860298,
+      "grad_norm": 2.4980483055114746,
+      "learning_rate": 3.3326988960791778e-06,
+      "loss": 1.3915,
+      "step": 1184
+    },
+    {
+      "epoch": 1.2172573189522342,
+      "grad_norm": 2.403468608856201,
+      "learning_rate": 3.330795584316711e-06,
+      "loss": 1.3239,
+      "step": 1185
+    },
+    {
+      "epoch": 1.2182845403184386,
+      "grad_norm": 2.6861846446990967,
+      "learning_rate": 3.3288922725542445e-06,
+      "loss": 1.4013,
+      "step": 1186
+    },
+    {
+      "epoch": 1.219311761684643,
+      "grad_norm": 2.525033712387085,
+      "learning_rate": 3.326988960791778e-06,
+      "loss": 1.4135,
+      "step": 1187
+    },
+    {
+      "epoch": 1.2203389830508475,
+      "grad_norm": 2.6705894470214844,
+      "learning_rate": 3.3250856490293117e-06,
+      "loss": 1.4587,
+      "step": 1188
+    },
+    {
+      "epoch": 1.221366204417052,
+      "grad_norm": 2.6108217239379883,
+      "learning_rate": 3.323182337266845e-06,
+      "loss": 1.4482,
+      "step": 1189
+    },
+    {
+      "epoch": 1.2223934257832563,
+      "grad_norm": 2.5227413177490234,
+      "learning_rate": 3.321279025504378e-06,
+      "loss": 1.432,
+      "step": 1190
+    },
+    {
+      "epoch": 1.2223934257832563,
+      "eval_loss": 1.756821870803833,
+      "eval_runtime": 37.9929,
+      "eval_samples_per_second": 26.321,
+      "eval_steps_per_second": 1.105,
+      "step": 1190
+    },
+    {
+      "epoch": 1.2234206471494606,
+      "grad_norm": 2.469766139984131,
+      "learning_rate": 3.3193757137419113e-06,
+      "loss": 1.4215,
+      "step": 1191
+    },
+    {
+      "epoch": 1.2244478685156652,
+      "grad_norm": 2.5257861614227295,
+      "learning_rate": 3.3174724019794447e-06,
+      "loss": 1.4624,
+      "step": 1192
+    },
+    {
+      "epoch": 1.2254750898818696,
+      "grad_norm": 2.5281760692596436,
+      "learning_rate": 3.315569090216978e-06,
+      "loss": 1.3799,
+      "step": 1193
+    },
+    {
+      "epoch": 1.226502311248074,
+      "grad_norm": 2.701521396636963,
+      "learning_rate": 3.3136657784545114e-06,
+      "loss": 1.4631,
+      "step": 1194
+    },
+    {
+      "epoch": 1.2275295326142783,
+      "grad_norm": 2.4239444732666016,
+      "learning_rate": 3.3117624666920443e-06,
+      "loss": 1.3746,
+      "step": 1195
+    },
+    {
+      "epoch": 1.2285567539804827,
+      "grad_norm": 2.430377721786499,
+      "learning_rate": 3.3098591549295777e-06,
+      "loss": 1.3584,
+      "step": 1196
+    },
+    {
+      "epoch": 1.2295839753466873,
+      "grad_norm": 2.7177658081054688,
+      "learning_rate": 3.307955843167111e-06,
+      "loss": 1.4301,
+      "step": 1197
+    },
+    {
+      "epoch": 1.2306111967128917,
+      "grad_norm": 2.512943983078003,
+      "learning_rate": 3.3060525314046444e-06,
+      "loss": 1.4111,
+      "step": 1198
+    },
+    {
+      "epoch": 1.231638418079096,
+      "grad_norm": 2.5269784927368164,
+      "learning_rate": 3.304149219642178e-06,
+      "loss": 1.4096,
+      "step": 1199
+    },
+    {
+      "epoch": 1.2326656394453004,
+      "grad_norm": 2.476409673690796,
+      "learning_rate": 3.3022459078797107e-06,
+      "loss": 1.4322,
+      "step": 1200
+    },
+    {
+      "epoch": 1.2326656394453004,
+      "eval_loss": 1.7645230293273926,
+      "eval_runtime": 37.9898,
+      "eval_samples_per_second": 26.323,
+      "eval_steps_per_second": 1.106,
+      "step": 1200
+    },
+    {
+      "epoch": 1.2336928608115048,
+      "grad_norm": 2.5339248180389404,
+      "learning_rate": 3.300342596117244e-06,
+      "loss": 1.4093,
+      "step": 1201
+    },
+    {
+      "epoch": 1.2347200821777093,
+      "grad_norm": 2.3505656719207764,
+      "learning_rate": 3.2984392843547775e-06,
+      "loss": 1.3623,
+      "step": 1202
+    },
+    {
+      "epoch": 1.2357473035439137,
+      "grad_norm": 2.400259256362915,
+      "learning_rate": 3.296535972592311e-06,
+      "loss": 1.3512,
+      "step": 1203
+    },
+    {
+      "epoch": 1.236774524910118,
+      "grad_norm": 2.561558723449707,
+      "learning_rate": 3.294632660829844e-06,
+      "loss": 1.3946,
+      "step": 1204
+    },
+    {
+      "epoch": 1.2378017462763227,
+      "grad_norm": 2.623100519180298,
+      "learning_rate": 3.292729349067377e-06,
+      "loss": 1.3972,
+      "step": 1205
+    },
+    {
+      "epoch": 1.238828967642527,
+      "grad_norm": 2.4889469146728516,
+      "learning_rate": 3.2908260373049105e-06,
+      "loss": 1.381,
+      "step": 1206
+    },
+    {
+      "epoch": 1.2398561890087314,
+      "grad_norm": 2.457334518432617,
+      "learning_rate": 3.288922725542444e-06,
+      "loss": 1.4345,
+      "step": 1207
+    },
+    {
+      "epoch": 1.2408834103749358,
+      "grad_norm": 2.440988540649414,
+      "learning_rate": 3.2870194137799776e-06,
+      "loss": 1.3901,
+      "step": 1208
+    },
+    {
+      "epoch": 1.2419106317411401,
+      "grad_norm": 2.4992308616638184,
+      "learning_rate": 3.285116102017511e-06,
+      "loss": 1.3694,
+      "step": 1209
+    },
+    {
+      "epoch": 1.2429378531073447,
+      "grad_norm": 2.611128091812134,
+      "learning_rate": 3.2832127902550444e-06,
+      "loss": 1.4344,
+      "step": 1210
+    },
+    {
+      "epoch": 1.2429378531073447,
+      "eval_loss": 1.763407826423645,
+      "eval_runtime": 37.7142,
+      "eval_samples_per_second": 26.515,
+      "eval_steps_per_second": 1.114,
+      "step": 1210
+    },
+    {
+      "epoch": 1.243965074473549,
+      "grad_norm": 2.6573522090911865,
+      "learning_rate": 3.2813094784925777e-06,
+      "loss": 1.4204,
+      "step": 1211
+    },
+    {
+      "epoch": 1.2449922958397535,
+      "grad_norm": 2.4785778522491455,
+      "learning_rate": 3.2794061667301107e-06,
+      "loss": 1.3617,
+      "step": 1212
+    },
+    {
+      "epoch": 1.2460195172059578,
+      "grad_norm": 2.4831974506378174,
+      "learning_rate": 3.277502854967644e-06,
+      "loss": 1.4004,
+      "step": 1213
+    },
+    {
+      "epoch": 1.2470467385721622,
+      "grad_norm": 2.6142327785491943,
+      "learning_rate": 3.2755995432051774e-06,
+      "loss": 1.3881,
+      "step": 1214
+    },
+    {
+      "epoch": 1.2480739599383668,
+      "grad_norm": 2.447962999343872,
+      "learning_rate": 3.2736962314427107e-06,
+      "loss": 1.3984,
+      "step": 1215
+    },
+    {
+      "epoch": 1.2491011813045712,
+      "grad_norm": 2.3895885944366455,
+      "learning_rate": 3.271792919680244e-06,
+      "loss": 1.4099,
+      "step": 1216
+    },
+    {
+      "epoch": 1.2501284026707755,
+      "grad_norm": 2.599364995956421,
+      "learning_rate": 3.269889607917777e-06,
+      "loss": 1.4575,
+      "step": 1217
+    },
+    {
+      "epoch": 1.25115562403698,
+      "grad_norm": 2.6294758319854736,
+      "learning_rate": 3.2679862961553104e-06,
+      "loss": 1.3752,
+      "step": 1218
+    },
+    {
+      "epoch": 1.2521828454031843,
+      "grad_norm": 2.50923228263855,
+      "learning_rate": 3.2660829843928438e-06,
+      "loss": 1.4017,
+      "step": 1219
+    },
+    {
+      "epoch": 1.2532100667693888,
+      "grad_norm": 2.626434564590454,
+      "learning_rate": 3.264179672630377e-06,
+      "loss": 1.4505,
+      "step": 1220
+    },
+    {
+      "epoch": 1.2532100667693888,
+      "eval_loss": 1.7672324180603027,
+      "eval_runtime": 37.8605,
+      "eval_samples_per_second": 26.413,
+      "eval_steps_per_second": 1.109,
+      "step": 1220
+    },
+    {
+      "epoch": 1.2542372881355932,
+      "grad_norm": 2.541663885116577,
+      "learning_rate": 3.2622763608679105e-06,
+      "loss": 1.4129,
+      "step": 1221
+    },
+    {
+      "epoch": 1.2552645095017976,
+      "grad_norm": 2.559877395629883,
+      "learning_rate": 3.2603730491054434e-06,
+      "loss": 1.3793,
+      "step": 1222
+    },
+    {
+      "epoch": 1.2562917308680022,
+      "grad_norm": 2.4085795879364014,
+      "learning_rate": 3.258469737342977e-06,
+      "loss": 1.3895,
+      "step": 1223
+    },
+    {
+      "epoch": 1.2573189522342065,
+      "grad_norm": 2.2942678928375244,
+      "learning_rate": 3.25656642558051e-06,
+      "loss": 1.3713,
+      "step": 1224
+    },
+    {
+      "epoch": 1.258346173600411,
+      "grad_norm": 2.561086893081665,
+      "learning_rate": 3.2546631138180435e-06,
+      "loss": 1.3835,
+      "step": 1225
+    },
+    {
+      "epoch": 1.2593733949666153,
+      "grad_norm": 2.4947667121887207,
+      "learning_rate": 3.252759802055577e-06,
+      "loss": 1.3793,
+      "step": 1226
+    },
+    {
+      "epoch": 1.2604006163328196,
+      "grad_norm": 2.6293253898620605,
+      "learning_rate": 3.25085649029311e-06,
+      "loss": 1.3935,
+      "step": 1227
+    },
+    {
+      "epoch": 1.2614278376990242,
+      "grad_norm": 2.5461983680725098,
+      "learning_rate": 3.248953178530644e-06,
+      "loss": 1.4201,
+      "step": 1228
+    },
+    {
+      "epoch": 1.2624550590652286,
+      "grad_norm": 2.381537675857544,
+      "learning_rate": 3.247049866768177e-06,
+      "loss": 1.3932,
+      "step": 1229
+    },
+    {
+      "epoch": 1.263482280431433,
+      "grad_norm": 2.381558418273926,
+      "learning_rate": 3.2451465550057103e-06,
+      "loss": 1.3425,
+      "step": 1230
+    },
+    {
+      "epoch": 1.263482280431433,
+      "eval_loss": 1.7713288068771362,
+      "eval_runtime": 37.9629,
+      "eval_samples_per_second": 26.342,
+      "eval_steps_per_second": 1.106,
+      "step": 1230
+    },
+    {
+      "epoch": 1.2645095017976373,
+      "grad_norm": 2.5877468585968018,
+      "learning_rate": 3.2432432432432437e-06,
+      "loss": 1.3917,
+      "step": 1231
+    },
+    {
+      "epoch": 1.2655367231638417,
+      "grad_norm": 2.6057848930358887,
+      "learning_rate": 3.241339931480777e-06,
+      "loss": 1.4217,
+      "step": 1232
+    },
+    {
+      "epoch": 1.2665639445300463,
+      "grad_norm": 2.5157413482666016,
+      "learning_rate": 3.2394366197183104e-06,
+      "loss": 1.427,
+      "step": 1233
+    },
+    {
+      "epoch": 1.2675911658962506,
+      "grad_norm": 2.4153692722320557,
+      "learning_rate": 3.2375333079558434e-06,
+      "loss": 1.3822,
+      "step": 1234
+    },
+    {
+      "epoch": 1.268618387262455,
+      "grad_norm": 2.5459582805633545,
+      "learning_rate": 3.2356299961933767e-06,
+      "loss": 1.4152,
+      "step": 1235
+    },
+    {
+      "epoch": 1.2696456086286596,
+      "grad_norm": 2.606468915939331,
+      "learning_rate": 3.23372668443091e-06,
+      "loss": 1.4278,
+      "step": 1236
+    },
+    {
+      "epoch": 1.270672829994864,
+      "grad_norm": 2.5061633586883545,
+      "learning_rate": 3.2318233726684435e-06,
+      "loss": 1.4512,
+      "step": 1237
+    },
+    {
+      "epoch": 1.2717000513610683,
+      "grad_norm": 2.3693087100982666,
+      "learning_rate": 3.229920060905977e-06,
+      "loss": 1.3791,
+      "step": 1238
+    },
+    {
+      "epoch": 1.2727272727272727,
+      "grad_norm": 2.479583978652954,
+      "learning_rate": 3.2280167491435098e-06,
+      "loss": 1.3814,
+      "step": 1239
+    },
+    {
+      "epoch": 1.273754494093477,
+      "grad_norm": 2.3258883953094482,
+      "learning_rate": 3.226113437381043e-06,
+      "loss": 1.3904,
+      "step": 1240
+    },
+    {
+      "epoch": 1.273754494093477,
+      "eval_loss": 1.7602728605270386,
+      "eval_runtime": 37.9604,
+      "eval_samples_per_second": 26.343,
+      "eval_steps_per_second": 1.106,
+      "step": 1240
+    },
+    {
+      "epoch": 1.2747817154596817,
+      "grad_norm": 2.4170308113098145,
+      "learning_rate": 3.2242101256185765e-06,
+      "loss": 1.4045,
+      "step": 1241
+    },
+    {
+      "epoch": 1.275808936825886,
+      "grad_norm": 2.578066825866699,
+      "learning_rate": 3.22230681385611e-06,
+      "loss": 1.4556,
+      "step": 1242
+    },
+    {
+      "epoch": 1.2768361581920904,
+      "grad_norm": 2.558187246322632,
+      "learning_rate": 3.220403502093643e-06,
+      "loss": 1.4036,
+      "step": 1243
+    },
+    {
+      "epoch": 1.2778633795582948,
+      "grad_norm": 2.598916530609131,
+      "learning_rate": 3.218500190331176e-06,
+      "loss": 1.4462,
+      "step": 1244
+    },
+    {
+      "epoch": 1.2788906009244991,
+      "grad_norm": 2.4528257846832275,
+      "learning_rate": 3.2165968785687095e-06,
+      "loss": 1.3855,
+      "step": 1245
+    },
+    {
+      "epoch": 1.2799178222907037,
+      "grad_norm": 2.5937459468841553,
+      "learning_rate": 3.214693566806243e-06,
+      "loss": 1.4191,
+      "step": 1246
+    },
+    {
+      "epoch": 1.280945043656908,
+      "grad_norm": 2.5324676036834717,
+      "learning_rate": 3.2127902550437762e-06,
+      "loss": 1.4162,
+      "step": 1247
+    },
+    {
+      "epoch": 1.2819722650231125,
+      "grad_norm": 2.4849791526794434,
+      "learning_rate": 3.21088694328131e-06,
+      "loss": 1.4037,
+      "step": 1248
+    },
+    {
+      "epoch": 1.2829994863893168,
+      "grad_norm": 2.389582633972168,
+      "learning_rate": 3.2089836315188434e-06,
+      "loss": 1.3966,
+      "step": 1249
+    },
+    {
+      "epoch": 1.2840267077555212,
+      "grad_norm": 2.387909173965454,
+      "learning_rate": 3.2070803197563767e-06,
+      "loss": 1.396,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2840267077555212,
+      "eval_loss": 1.7642556428909302,
+      "eval_runtime": 37.9446,
+      "eval_samples_per_second": 26.354,
+      "eval_steps_per_second": 1.107,
+      "step": 1250
+    },
+    {
+      "epoch": 1.2850539291217258,
+      "grad_norm": 2.5014612674713135,
+      "learning_rate": 3.2051770079939097e-06,
+      "loss": 1.4128,
+      "step": 1251
+    },
+    {
+      "epoch": 1.2860811504879301,
+      "grad_norm": 2.497950553894043,
+      "learning_rate": 3.203273696231443e-06,
+      "loss": 1.4412,
+      "step": 1252
+    },
+    {
+      "epoch": 1.2871083718541345,
+      "grad_norm": 2.4371297359466553,
+      "learning_rate": 3.2013703844689764e-06,
+      "loss": 1.3754,
+      "step": 1253
+    },
+    {
+      "epoch": 1.288135593220339,
+      "grad_norm": 2.4336745738983154,
+      "learning_rate": 3.1994670727065098e-06,
+      "loss": 1.4258,
+      "step": 1254
+    },
+    {
+      "epoch": 1.2891628145865435,
+      "grad_norm": 2.5165536403656006,
+      "learning_rate": 3.197563760944043e-06,
+      "loss": 1.4414,
+      "step": 1255
+    },
+    {
+      "epoch": 1.2901900359527478,
+      "grad_norm": 2.5060458183288574,
+      "learning_rate": 3.195660449181576e-06,
+      "loss": 1.4336,
+      "step": 1256
+    },
+    {
+      "epoch": 1.2912172573189522,
+      "grad_norm": 2.584284543991089,
+      "learning_rate": 3.1937571374191094e-06,
+      "loss": 1.3768,
+      "step": 1257
+    },
+    {
+      "epoch": 1.2922444786851566,
+      "grad_norm": 2.615629196166992,
+      "learning_rate": 3.191853825656643e-06,
+      "loss": 1.401,
+      "step": 1258
+    },
+    {
+      "epoch": 1.2932717000513612,
+      "grad_norm": 2.5989596843719482,
+      "learning_rate": 3.189950513894176e-06,
+      "loss": 1.3915,
+      "step": 1259
+    },
+    {
+      "epoch": 1.2942989214175655,
+      "grad_norm": 2.5507843494415283,
+      "learning_rate": 3.1880472021317095e-06,
+      "loss": 1.3979,
+      "step": 1260
+    },
+    {
+      "epoch": 1.2942989214175655,
+      "eval_loss": 1.7715612649917603,
+      "eval_runtime": 37.9859,
+      "eval_samples_per_second": 26.326,
+      "eval_steps_per_second": 1.106,
+      "step": 1260
+    },
+    {
+      "epoch": 1.29532614278377,
+      "grad_norm": 2.512140989303589,
+      "learning_rate": 3.1861438903692425e-06,
+      "loss": 1.3523,
+      "step": 1261
+    },
+    {
+      "epoch": 1.2963533641499743,
+      "grad_norm": 2.4311885833740234,
+      "learning_rate": 3.184240578606776e-06,
+      "loss": 1.3828,
+      "step": 1262
+    },
+    {
+      "epoch": 1.2973805855161786,
+      "grad_norm": 2.439265489578247,
+      "learning_rate": 3.182337266844309e-06,
+      "loss": 1.402,
+      "step": 1263
+    },
+    {
+      "epoch": 1.2984078068823832,
+      "grad_norm": 2.5773651599884033,
+      "learning_rate": 3.1804339550818426e-06,
+      "loss": 1.4362,
+      "step": 1264
+    },
+    {
+      "epoch": 1.2994350282485876,
+      "grad_norm": 2.456021547317505,
+      "learning_rate": 3.178530643319376e-06,
+      "loss": 1.3918,
+      "step": 1265
+    },
+    {
+      "epoch": 1.300462249614792,
+      "grad_norm": 2.44404673576355,
+      "learning_rate": 3.176627331556909e-06,
+      "loss": 1.3662,
+      "step": 1266
+    },
+    {
+      "epoch": 1.3014894709809965,
+      "grad_norm": 2.778027057647705,
+      "learning_rate": 3.1747240197944422e-06,
+      "loss": 1.4112,
+      "step": 1267
+    },
+    {
+      "epoch": 1.3025166923472007,
+      "grad_norm": 2.476496934890747,
+      "learning_rate": 3.172820708031976e-06,
+      "loss": 1.3654,
+      "step": 1268
+    },
+    {
+      "epoch": 1.3035439137134053,
+      "grad_norm": 2.6415998935699463,
+      "learning_rate": 3.1709173962695094e-06,
+      "loss": 1.4441,
+      "step": 1269
+    },
+    {
+      "epoch": 1.3045711350796096,
+      "grad_norm": 2.5046234130859375,
+      "learning_rate": 3.1690140845070427e-06,
+      "loss": 1.4316,
+      "step": 1270
+    },
+    {
+      "epoch": 1.3045711350796096,
+      "eval_loss": 1.7736135721206665,
+      "eval_runtime": 37.9392,
+      "eval_samples_per_second": 26.358,
+      "eval_steps_per_second": 1.107,
+      "step": 1270
+    },
+    {
+      "epoch": 1.305598356445814,
+      "grad_norm": 2.6234474182128906,
+      "learning_rate": 3.167110772744576e-06,
+      "loss": 1.4,
+      "step": 1271
+    },
+    {
+      "epoch": 1.3066255778120186,
+      "grad_norm": 2.488297700881958,
+      "learning_rate": 3.1652074609821095e-06,
+      "loss": 1.3908,
+      "step": 1272
+    },
+    {
+      "epoch": 1.307652799178223,
+      "grad_norm": 2.3759512901306152,
+      "learning_rate": 3.1633041492196424e-06,
+      "loss": 1.3908,
+      "step": 1273
+    },
+    {
+      "epoch": 1.3086800205444273,
+      "grad_norm": 2.487281084060669,
+      "learning_rate": 3.1614008374571758e-06,
+      "loss": 1.4217,
+      "step": 1274
+    },
+    {
+      "epoch": 1.3097072419106317,
+      "grad_norm": 2.5366780757904053,
+      "learning_rate": 3.159497525694709e-06,
+      "loss": 1.4229,
+      "step": 1275
+    },
+    {
+      "epoch": 1.310734463276836,
+      "grad_norm": 2.416141986846924,
+      "learning_rate": 3.1575942139322425e-06,
+      "loss": 1.3865,
+      "step": 1276
+    },
+    {
+      "epoch": 1.3117616846430407,
+      "grad_norm": 2.537964344024658,
+      "learning_rate": 3.155690902169776e-06,
+      "loss": 1.4141,
+      "step": 1277
+    },
+    {
+      "epoch": 1.312788906009245,
+      "grad_norm": 2.481692314147949,
+      "learning_rate": 3.1537875904073088e-06,
+      "loss": 1.383,
+      "step": 1278
+    },
+    {
+      "epoch": 1.3138161273754494,
+      "grad_norm": 2.4898593425750732,
+      "learning_rate": 3.151884278644842e-06,
+      "loss": 1.4078,
+      "step": 1279
+    },
+    {
+      "epoch": 1.3148433487416538,
+      "grad_norm": 2.4692323207855225,
+      "learning_rate": 3.1499809668823755e-06,
+      "loss": 1.3818,
+      "step": 1280
+    },
+    {
+      "epoch": 1.3148433487416538,
+      "eval_loss": 1.7660051584243774,
+      "eval_runtime": 37.8922,
+      "eval_samples_per_second": 26.391,
+      "eval_steps_per_second": 1.108,
+      "step": 1280
+    },
+    {
+      "epoch": 1.3158705701078581,
+      "grad_norm": 2.475862503051758,
+      "learning_rate": 3.148077655119909e-06,
+      "loss": 1.4384,
+      "step": 1281
+    },
+    {
+      "epoch": 1.3168977914740627,
+      "grad_norm": 2.2839648723602295,
+      "learning_rate": 3.1461743433574422e-06,
+      "loss": 1.3722,
+      "step": 1282
+    },
+    {
+      "epoch": 1.317925012840267,
+      "grad_norm": 2.3965585231781006,
+      "learning_rate": 3.144271031594975e-06,
+      "loss": 1.4262,
+      "step": 1283
+    },
+    {
+      "epoch": 1.3189522342064715,
+      "grad_norm": 2.4107236862182617,
+      "learning_rate": 3.1423677198325085e-06,
+      "loss": 1.4216,
+      "step": 1284
+    },
+    {
+      "epoch": 1.319979455572676,
+      "grad_norm": 2.438899040222168,
+      "learning_rate": 3.140464408070042e-06,
+      "loss": 1.383,
+      "step": 1285
+    },
+    {
+      "epoch": 1.3210066769388804,
+      "grad_norm": 2.4146060943603516,
+      "learning_rate": 3.1385610963075753e-06,
+      "loss": 1.3827,
+      "step": 1286
+    },
+    {
+      "epoch": 1.3220338983050848,
+      "grad_norm": 2.4193496704101562,
+      "learning_rate": 3.1366577845451086e-06,
+      "loss": 1.3707,
+      "step": 1287
+    },
+    {
+      "epoch": 1.3230611196712891,
+      "grad_norm": 2.5284323692321777,
+      "learning_rate": 3.134754472782642e-06,
+      "loss": 1.3623,
+      "step": 1288
+    },
+    {
+      "epoch": 1.3240883410374935,
+      "grad_norm": 2.421509265899658,
+      "learning_rate": 3.1328511610201758e-06,
+      "loss": 1.35,
+      "step": 1289
+    },
+    {
+      "epoch": 1.325115562403698,
+      "grad_norm": 2.3964431285858154,
+      "learning_rate": 3.1309478492577087e-06,
+      "loss": 1.3486,
+      "step": 1290
+    },
+    {
+      "epoch": 1.325115562403698,
+      "eval_loss": 1.7699168920516968,
+      "eval_runtime": 37.9003,
+      "eval_samples_per_second": 26.385,
+      "eval_steps_per_second": 1.108,
+      "step": 1290
+    },
+    {
+      "epoch": 1.3261427837699025,
+      "grad_norm": 2.4371817111968994,
+      "learning_rate": 3.129044537495242e-06,
+      "loss": 1.3958,
+      "step": 1291
+    },
+    {
+      "epoch": 1.3271700051361068,
+      "grad_norm": 2.45233154296875,
+      "learning_rate": 3.1271412257327754e-06,
+      "loss": 1.4175,
+      "step": 1292
+    },
+    {
+      "epoch": 1.3281972265023112,
+      "grad_norm": 2.314384937286377,
+      "learning_rate": 3.125237913970309e-06,
+      "loss": 1.3701,
+      "step": 1293
+    },
+    {
+      "epoch": 1.3292244478685156,
+      "grad_norm": 2.5603694915771484,
+      "learning_rate": 3.123334602207842e-06,
+      "loss": 1.3489,
+      "step": 1294
+    },
+    {
+      "epoch": 1.3302516692347202,
+      "grad_norm": 2.5919623374938965,
+      "learning_rate": 3.121431290445375e-06,
+      "loss": 1.4472,
+      "step": 1295
+    },
+    {
+      "epoch": 1.3312788906009245,
+      "grad_norm": 2.6912758350372314,
+      "learning_rate": 3.1195279786829085e-06,
+      "loss": 1.4917,
+      "step": 1296
+    },
+    {
+      "epoch": 1.332306111967129,
+      "grad_norm": 2.439089298248291,
+      "learning_rate": 3.117624666920442e-06,
+      "loss": 1.3776,
+      "step": 1297
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 2.5207953453063965,
+      "learning_rate": 3.115721355157975e-06,
+      "loss": 1.3784,
+      "step": 1298
+    },
+    {
+      "epoch": 1.3343605546995376,
+      "grad_norm": 2.507791519165039,
+      "learning_rate": 3.1138180433955085e-06,
+      "loss": 1.3346,
+      "step": 1299
+    },
+    {
+      "epoch": 1.3353877760657422,
+      "grad_norm": 2.450531482696533,
+      "learning_rate": 3.1119147316330415e-06,
+      "loss": 1.4151,
+      "step": 1300
+    },
+    {
+      "epoch": 1.3353877760657422,
+      "eval_loss": 1.7761608362197876,
+      "eval_runtime": 37.9041,
+      "eval_samples_per_second": 26.382,
+      "eval_steps_per_second": 1.108,
+      "step": 1300
+    },
+    {
+      "epoch": 1.3364149974319466,
+      "grad_norm": 2.488255262374878,
+      "learning_rate": 3.110011419870575e-06,
+      "loss": 1.4475,
+      "step": 1301
+    },
+    {
+      "epoch": 1.337442218798151,
+      "grad_norm": 2.5059807300567627,
+      "learning_rate": 3.1081081081081082e-06,
+      "loss": 1.3812,
+      "step": 1302
+    },
+    {
+      "epoch": 1.3384694401643555,
+      "grad_norm": 2.5579488277435303,
+      "learning_rate": 3.1062047963456416e-06,
+      "loss": 1.4134,
+      "step": 1303
+    },
+    {
+      "epoch": 1.33949666153056,
+      "grad_norm": 2.6170148849487305,
+      "learning_rate": 3.104301484583175e-06,
+      "loss": 1.459,
+      "step": 1304
+    },
+    {
+      "epoch": 1.3405238828967643,
+      "grad_norm": 2.5572125911712646,
+      "learning_rate": 3.1023981728207083e-06,
+      "loss": 1.4354,
+      "step": 1305
+    },
+    {
+      "epoch": 1.3415511042629686,
+      "grad_norm": 2.396883487701416,
+      "learning_rate": 3.1004948610582412e-06,
+      "loss": 1.4016,
+      "step": 1306
+    },
+    {
+      "epoch": 1.342578325629173,
+      "grad_norm": 2.5869390964508057,
+      "learning_rate": 3.0985915492957746e-06,
+      "loss": 1.3945,
+      "step": 1307
+    },
+    {
+      "epoch": 1.3436055469953776,
+      "grad_norm": 2.4271085262298584,
+      "learning_rate": 3.096688237533308e-06,
+      "loss": 1.3506,
+      "step": 1308
+    },
+    {
+      "epoch": 1.344632768361582,
+      "grad_norm": 2.398407220840454,
+      "learning_rate": 3.0947849257708418e-06,
+      "loss": 1.4475,
+      "step": 1309
+    },
+    {
+      "epoch": 1.3456599897277863,
+      "grad_norm": 2.441934823989868,
+      "learning_rate": 3.092881614008375e-06,
+      "loss": 1.3972,
+      "step": 1310
+    },
+    {
+      "epoch": 1.3456599897277863,
+      "eval_loss": 1.774910569190979,
+      "eval_runtime": 37.6105,
+      "eval_samples_per_second": 26.588,
+      "eval_steps_per_second": 1.117,
+      "step": 1310
+    },
+    {
+      "epoch": 1.3466872110939907,
+      "grad_norm": 2.5015370845794678,
+      "learning_rate": 3.0909783022459085e-06,
+      "loss": 1.4418,
+      "step": 1311
+    },
+    {
+      "epoch": 1.347714432460195,
+      "grad_norm": 2.4616997241973877,
+      "learning_rate": 3.0890749904834414e-06,
+      "loss": 1.3833,
+      "step": 1312
+    },
+    {
+      "epoch": 1.3487416538263997,
+      "grad_norm": 2.5993902683258057,
+      "learning_rate": 3.0871716787209748e-06,
+      "loss": 1.4385,
+      "step": 1313
+    },
+    {
+      "epoch": 1.349768875192604,
+      "grad_norm": 2.518195867538452,
+      "learning_rate": 3.085268366958508e-06,
+      "loss": 1.4127,
+      "step": 1314
+    },
+    {
+      "epoch": 1.3507960965588084,
+      "grad_norm": 2.545189380645752,
+      "learning_rate": 3.0833650551960415e-06,
+      "loss": 1.42,
+      "step": 1315
+    },
+    {
+      "epoch": 1.3518233179250128,
+      "grad_norm": 2.461818218231201,
+      "learning_rate": 3.081461743433575e-06,
+      "loss": 1.3796,
+      "step": 1316
+    },
+    {
+      "epoch": 1.3528505392912171,
+      "grad_norm": 2.4081408977508545,
+      "learning_rate": 3.079558431671108e-06,
+      "loss": 1.3939,
+      "step": 1317
+    },
+    {
+      "epoch": 1.3538777606574217,
+      "grad_norm": 2.2885539531707764,
+      "learning_rate": 3.077655119908641e-06,
+      "loss": 1.3645,
+      "step": 1318
+    },
+    {
+      "epoch": 1.354904982023626,
+      "grad_norm": 2.453791618347168,
+      "learning_rate": 3.0757518081461745e-06,
+      "loss": 1.3906,
+      "step": 1319
+    },
+    {
+      "epoch": 1.3559322033898304,
+      "grad_norm": 2.4538352489471436,
+      "learning_rate": 3.073848496383708e-06,
+      "loss": 1.4127,
+      "step": 1320
+    },
+    {
+      "epoch": 1.3559322033898304,
+      "eval_loss": 1.7821755409240723,
+      "eval_runtime": 37.7585,
+      "eval_samples_per_second": 26.484,
+      "eval_steps_per_second": 1.112,
+      "step": 1320
+    },
+    {
+      "epoch": 1.356959424756035,
+      "grad_norm": 2.438425064086914,
+      "learning_rate": 3.0719451846212413e-06,
+      "loss": 1.3763,
+      "step": 1321
+    },
+    {
+      "epoch": 1.3579866461222394,
+      "grad_norm": 2.4398279190063477,
+      "learning_rate": 3.0700418728587746e-06,
+      "loss": 1.3963,
+      "step": 1322
+    },
+    {
+      "epoch": 1.3590138674884438,
+      "grad_norm": 2.6763861179351807,
+      "learning_rate": 3.0681385610963076e-06,
+      "loss": 1.3837,
+      "step": 1323
+    },
+    {
+      "epoch": 1.3600410888546481,
+      "grad_norm": 2.685056209564209,
+      "learning_rate": 3.066235249333841e-06,
+      "loss": 1.4158,
+      "step": 1324
+    },
+    {
+      "epoch": 1.3610683102208525,
+      "grad_norm": 2.6663355827331543,
+      "learning_rate": 3.0643319375713743e-06,
+      "loss": 1.4533,
+      "step": 1325
+    },
+    {
+      "epoch": 1.362095531587057,
+      "grad_norm": 2.4613428115844727,
+      "learning_rate": 3.0624286258089076e-06,
+      "loss": 1.3927,
+      "step": 1326
+    },
+    {
+      "epoch": 1.3631227529532615,
+      "grad_norm": 2.348334789276123,
+      "learning_rate": 3.060525314046441e-06,
+      "loss": 1.3782,
+      "step": 1327
+    },
+    {
+      "epoch": 1.3641499743194658,
+      "grad_norm": 2.5011332035064697,
+      "learning_rate": 3.058622002283974e-06,
+      "loss": 1.4175,
+      "step": 1328
+    },
+    {
+      "epoch": 1.3651771956856702,
+      "grad_norm": 2.4692256450653076,
+      "learning_rate": 3.0567186905215077e-06,
+      "loss": 1.4319,
+      "step": 1329
+    },
+    {
+      "epoch": 1.3662044170518746,
+      "grad_norm": 2.616990089416504,
+      "learning_rate": 3.054815378759041e-06,
+      "loss": 1.4264,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3662044170518746,
+      "eval_loss": 1.781583309173584,
+      "eval_runtime": 37.9291,
+      "eval_samples_per_second": 26.365,
+      "eval_steps_per_second": 1.107,
+      "step": 1330
+    },
+    {
+      "epoch": 1.3672316384180792,
+      "grad_norm": 2.5416364669799805,
+      "learning_rate": 3.0529120669965745e-06,
+      "loss": 1.4249,
+      "step": 1331
+    },
+    {
+      "epoch": 1.3682588597842835,
+      "grad_norm": 2.5346689224243164,
+      "learning_rate": 3.051008755234108e-06,
+      "loss": 1.3647,
+      "step": 1332
+    },
+    {
+      "epoch": 1.3692860811504879,
+      "grad_norm": 2.585714340209961,
+      "learning_rate": 3.049105443471641e-06,
+      "loss": 1.3714,
+      "step": 1333
+    },
+    {
+      "epoch": 1.3703133025166925,
+      "grad_norm": 2.5958940982818604,
+      "learning_rate": 3.0472021317091745e-06,
+      "loss": 1.5041,
+      "step": 1334
+    },
+    {
+      "epoch": 1.3713405238828968,
+      "grad_norm": 2.4795925617218018,
+      "learning_rate": 3.0452988199467075e-06,
+      "loss": 1.4317,
+      "step": 1335
+    },
+    {
+      "epoch": 1.3723677452491012,
+      "grad_norm": 2.4608285427093506,
+      "learning_rate": 3.043395508184241e-06,
+      "loss": 1.4259,
+      "step": 1336
+    },
+    {
+      "epoch": 1.3733949666153056,
+      "grad_norm": 2.5970380306243896,
+      "learning_rate": 3.041492196421774e-06,
+      "loss": 1.3859,
+      "step": 1337
+    },
+    {
+      "epoch": 1.37442218798151,
+      "grad_norm": 2.517047166824341,
+      "learning_rate": 3.0395888846593076e-06,
+      "loss": 1.4086,
+      "step": 1338
+    },
+    {
+      "epoch": 1.3754494093477145,
+      "grad_norm": 2.4541358947753906,
+      "learning_rate": 3.037685572896841e-06,
+      "loss": 1.3959,
+      "step": 1339
+    },
+    {
+      "epoch": 1.376476630713919,
+      "grad_norm": 2.6143250465393066,
+      "learning_rate": 3.035782261134374e-06,
+      "loss": 1.4088,
+      "step": 1340
+    },
+    {
+      "epoch": 1.376476630713919,
+      "eval_loss": 1.7629026174545288,
+      "eval_runtime": 38.0156,
+      "eval_samples_per_second": 26.305,
+      "eval_steps_per_second": 1.105,
+      "step": 1340
+    },
+    {
+      "epoch": 1.3775038520801233,
+      "grad_norm": 2.763033151626587,
+      "learning_rate": 3.0338789493719072e-06,
+      "loss": 1.4104,
+      "step": 1341
+    },
+    {
+      "epoch": 1.3785310734463276,
+      "grad_norm": 2.523888349533081,
+      "learning_rate": 3.0319756376094406e-06,
+      "loss": 1.4073,
+      "step": 1342
+    },
+    {
+      "epoch": 1.379558294812532,
+      "grad_norm": 2.5393309593200684,
+      "learning_rate": 3.030072325846974e-06,
+      "loss": 1.4145,
+      "step": 1343
+    },
+    {
+      "epoch": 1.3805855161787366,
+      "grad_norm": 2.455733060836792,
+      "learning_rate": 3.0281690140845073e-06,
+      "loss": 1.3782,
+      "step": 1344
+    },
+    {
+      "epoch": 1.381612737544941,
+      "grad_norm": 2.600311517715454,
+      "learning_rate": 3.0262657023220403e-06,
+      "loss": 1.4158,
+      "step": 1345
+    },
+    {
+      "epoch": 1.3826399589111453,
+      "grad_norm": 2.453249931335449,
+      "learning_rate": 3.0243623905595736e-06,
+      "loss": 1.414,
+      "step": 1346
+    },
+    {
+      "epoch": 1.3836671802773497,
+      "grad_norm": 2.603515625,
+      "learning_rate": 3.022459078797107e-06,
+      "loss": 1.4102,
+      "step": 1347
+    },
+    {
+      "epoch": 1.384694401643554,
+      "grad_norm": 2.6515872478485107,
+      "learning_rate": 3.0205557670346404e-06,
+      "loss": 1.4063,
+      "step": 1348
+    },
+    {
+      "epoch": 1.3857216230097587,
+      "grad_norm": 2.412846565246582,
+      "learning_rate": 3.018652455272174e-06,
+      "loss": 1.3739,
+      "step": 1349
+    },
+    {
+      "epoch": 1.386748844375963,
+      "grad_norm": 2.748563051223755,
+      "learning_rate": 3.0167491435097075e-06,
+      "loss": 1.4909,
+      "step": 1350
+    },
+    {
+      "epoch": 1.386748844375963,
+      "eval_loss": 1.7653924226760864,
+      "eval_runtime": 37.9855,
+      "eval_samples_per_second": 26.326,
+      "eval_steps_per_second": 1.106,
+      "step": 1350
+    },
+    {
+      "epoch": 1.3877760657421674,
+      "grad_norm": 2.7861363887786865,
+      "learning_rate": 3.014845831747241e-06,
+      "loss": 1.4028,
+      "step": 1351
+    },
+    {
+      "epoch": 1.388803287108372,
+      "grad_norm": 2.5898189544677734,
+      "learning_rate": 3.012942519984774e-06,
+      "loss": 1.3793,
+      "step": 1352
+    },
+    {
+      "epoch": 1.3898305084745763,
+      "grad_norm": 2.4240355491638184,
+      "learning_rate": 3.011039208222307e-06,
+      "loss": 1.3889,
+      "step": 1353
+    },
+    {
+      "epoch": 1.3908577298407807,
+      "grad_norm": 2.4492504596710205,
+      "learning_rate": 3.0091358964598405e-06,
+      "loss": 1.3464,
+      "step": 1354
+    },
+    {
+      "epoch": 1.391884951206985,
+      "grad_norm": 2.4425718784332275,
+      "learning_rate": 3.007232584697374e-06,
+      "loss": 1.4073,
+      "step": 1355
+    },
+    {
+      "epoch": 1.3929121725731894,
+      "grad_norm": 2.5057594776153564,
+      "learning_rate": 3.0053292729349073e-06,
+      "loss": 1.4053,
+      "step": 1356
+    },
+    {
+      "epoch": 1.393939393939394,
+      "grad_norm": 2.5386273860931396,
+      "learning_rate": 3.00342596117244e-06,
+      "loss": 1.3991,
+      "step": 1357
+    },
+    {
+      "epoch": 1.3949666153055984,
+      "grad_norm": 2.6289730072021484,
+      "learning_rate": 3.0015226494099736e-06,
+      "loss": 1.3971,
+      "step": 1358
+    },
+    {
+      "epoch": 1.3959938366718028,
+      "grad_norm": 2.5091774463653564,
+      "learning_rate": 2.999619337647507e-06,
+      "loss": 1.3815,
+      "step": 1359
+    },
+    {
+      "epoch": 1.3970210580380071,
+      "grad_norm": 2.567659616470337,
+      "learning_rate": 2.9977160258850403e-06,
+      "loss": 1.3786,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3970210580380071,
+      "eval_loss": 1.7706127166748047,
+      "eval_runtime": 37.9897,
+      "eval_samples_per_second": 26.323,
+      "eval_steps_per_second": 1.106,
+      "step": 1360
+    },
+    {
+      "epoch": 1.3980482794042115,
+      "grad_norm": 2.5508553981781006,
+      "learning_rate": 2.9958127141225736e-06,
+      "loss": 1.3629,
+      "step": 1361
+    },
+    {
+      "epoch": 1.399075500770416,
+      "grad_norm": 2.624800205230713,
+      "learning_rate": 2.9939094023601066e-06,
+      "loss": 1.4355,
+      "step": 1362
+    },
+    {
+      "epoch": 1.4001027221366205,
+      "grad_norm": 2.432114839553833,
+      "learning_rate": 2.99200609059764e-06,
+      "loss": 1.3995,
+      "step": 1363
+    },
+    {
+      "epoch": 1.4011299435028248,
+      "grad_norm": 2.4633607864379883,
+      "learning_rate": 2.9901027788351733e-06,
+      "loss": 1.3499,
+      "step": 1364
+    },
+    {
+      "epoch": 1.4021571648690292,
+      "grad_norm": 2.557415246963501,
+      "learning_rate": 2.9881994670727067e-06,
+      "loss": 1.3614,
+      "step": 1365
+    },
+    {
+      "epoch": 1.4031843862352336,
+      "grad_norm": 2.411295175552368,
+      "learning_rate": 2.98629615531024e-06,
+      "loss": 1.3857,
+      "step": 1366
+    },
+    {
+      "epoch": 1.4042116076014381,
+      "grad_norm": 2.429797410964966,
+      "learning_rate": 2.984392843547773e-06,
+      "loss": 1.3701,
+      "step": 1367
+    },
+    {
+      "epoch": 1.4052388289676425,
+      "grad_norm": 2.6080522537231445,
+      "learning_rate": 2.9824895317853063e-06,
+      "loss": 1.3996,
+      "step": 1368
+    },
+    {
+      "epoch": 1.4062660503338469,
+      "grad_norm": 2.5128798484802246,
+      "learning_rate": 2.98058622002284e-06,
+      "loss": 1.3973,
+      "step": 1369
+    },
+    {
+      "epoch": 1.4072932717000515,
+      "grad_norm": 2.5906126499176025,
+      "learning_rate": 2.9786829082603735e-06,
+      "loss": 1.4412,
+      "step": 1370
+    },
+    {
+      "epoch": 1.4072932717000515,
+      "eval_loss": 1.7700779438018799,
+      "eval_runtime": 38.017,
+      "eval_samples_per_second": 26.304,
+      "eval_steps_per_second": 1.105,
+      "step": 1370
+    },
+    {
+      "epoch": 1.4083204930662558,
+      "grad_norm": 2.4890248775482178,
+      "learning_rate": 2.976779596497907e-06,
+      "loss": 1.4251,
+      "step": 1371
+    },
+    {
+      "epoch": 1.4093477144324602,
+      "grad_norm": 2.3803439140319824,
+      "learning_rate": 2.97487628473544e-06,
+      "loss": 1.4363,
+      "step": 1372
+    },
+    {
+      "epoch": 1.4103749357986646,
+      "grad_norm": 2.461958646774292,
+      "learning_rate": 2.9729729729729736e-06,
+      "loss": 1.3948,
+      "step": 1373
+    },
+    {
+      "epoch": 1.411402157164869,
+      "grad_norm": 2.419793128967285,
+      "learning_rate": 2.9710696612105065e-06,
+      "loss": 1.4367,
+      "step": 1374
+    },
+    {
+      "epoch": 1.4124293785310735,
+      "grad_norm": 2.5748846530914307,
+      "learning_rate": 2.96916634944804e-06,
+      "loss": 1.3873,
+      "step": 1375
+    },
+    {
+      "epoch": 1.413456599897278,
+      "grad_norm": 2.564558506011963,
+      "learning_rate": 2.9672630376855732e-06,
+      "loss": 1.4395,
+      "step": 1376
+    },
+    {
+      "epoch": 1.4144838212634823,
+      "grad_norm": 2.596085548400879,
+      "learning_rate": 2.9653597259231066e-06,
+      "loss": 1.4091,
+      "step": 1377
+    },
+    {
+      "epoch": 1.4155110426296866,
+      "grad_norm": 2.49029278755188,
+      "learning_rate": 2.96345641416064e-06,
+      "loss": 1.3639,
+      "step": 1378
+    },
+    {
+      "epoch": 1.416538263995891,
+      "grad_norm": 2.506887912750244,
+      "learning_rate": 2.961553102398173e-06,
+      "loss": 1.3065,
+      "step": 1379
+    },
+    {
+      "epoch": 1.4175654853620956,
+      "grad_norm": 2.517609119415283,
+      "learning_rate": 2.9596497906357063e-06,
+      "loss": 1.4365,
+      "step": 1380
+    },
+    {
+      "epoch": 1.4175654853620956,
+      "eval_loss": 1.7673696279525757,
+      "eval_runtime": 37.9941,
+      "eval_samples_per_second": 26.32,
+      "eval_steps_per_second": 1.105,
+      "step": 1380
+    },
+    {
+      "epoch": 1.4185927067283,
+      "grad_norm": 2.6870028972625732,
+      "learning_rate": 2.9577464788732396e-06,
+      "loss": 1.4386,
+      "step": 1381
+    },
+    {
+      "epoch": 1.4196199280945043,
+      "grad_norm": 2.3195550441741943,
+      "learning_rate": 2.955843167110773e-06,
+      "loss": 1.3771,
+      "step": 1382
+    },
+    {
+      "epoch": 1.420647149460709,
+      "grad_norm": 2.4306302070617676,
+      "learning_rate": 2.9539398553483064e-06,
+      "loss": 1.3963,
+      "step": 1383
+    },
+    {
+      "epoch": 1.4216743708269133,
+      "grad_norm": 2.4227781295776367,
+      "learning_rate": 2.9520365435858393e-06,
+      "loss": 1.394,
+      "step": 1384
+    },
+    {
+      "epoch": 1.4227015921931176,
+      "grad_norm": 2.521425485610962,
+      "learning_rate": 2.9501332318233727e-06,
+      "loss": 1.4664,
+      "step": 1385
+    },
+    {
+      "epoch": 1.423728813559322,
+      "grad_norm": 2.533363103866577,
+      "learning_rate": 2.948229920060906e-06,
+      "loss": 1.4072,
+      "step": 1386
+    },
+    {
+      "epoch": 1.4247560349255264,
+      "grad_norm": 2.378007173538208,
+      "learning_rate": 2.9463266082984394e-06,
+      "loss": 1.38,
+      "step": 1387
+    },
+    {
+      "epoch": 1.425783256291731,
+      "grad_norm": 2.476451873779297,
+      "learning_rate": 2.9444232965359727e-06,
+      "loss": 1.3864,
+      "step": 1388
+    },
+    {
+      "epoch": 1.4268104776579353,
+      "grad_norm": 2.428365707397461,
+      "learning_rate": 2.9425199847735057e-06,
+      "loss": 1.4081,
+      "step": 1389
+    },
+    {
+      "epoch": 1.4278376990241397,
+      "grad_norm": 2.4382991790771484,
+      "learning_rate": 2.94061667301104e-06,
+      "loss": 1.3806,
+      "step": 1390
+    },
+    {
+      "epoch": 1.4278376990241397,
+      "eval_loss": 1.7905043363571167,
+      "eval_runtime": 38.0264,
+      "eval_samples_per_second": 26.298,
+      "eval_steps_per_second": 1.104,
+      "step": 1390
+    },
+    {
+      "epoch": 1.428864920390344,
+      "grad_norm": 2.414433240890503,
+      "learning_rate": 2.938713361248573e-06,
+      "loss": 1.3113,
+      "step": 1391
+    },
+    {
+      "epoch": 1.4298921417565484,
+      "grad_norm": 2.450570821762085,
+      "learning_rate": 2.936810049486106e-06,
+      "loss": 1.4023,
+      "step": 1392
+    },
+    {
+      "epoch": 1.430919363122753,
+      "grad_norm": 2.4372148513793945,
+      "learning_rate": 2.9349067377236396e-06,
+      "loss": 1.4057,
+      "step": 1393
+    },
+    {
+      "epoch": 1.4319465844889574,
+      "grad_norm": 2.4278311729431152,
+      "learning_rate": 2.933003425961173e-06,
+      "loss": 1.3574,
+      "step": 1394
+    },
+    {
+      "epoch": 1.4329738058551618,
+      "grad_norm": 2.448517322540283,
+      "learning_rate": 2.9311001141987063e-06,
+      "loss": 1.3722,
+      "step": 1395
+    },
+    {
+      "epoch": 1.4340010272213661,
+      "grad_norm": 2.354310989379883,
+      "learning_rate": 2.9291968024362392e-06,
+      "loss": 1.3682,
+      "step": 1396
+    },
+    {
+      "epoch": 1.4350282485875705,
+      "grad_norm": 2.3318262100219727,
+      "learning_rate": 2.9272934906737726e-06,
+      "loss": 1.3604,
+      "step": 1397
+    },
+    {
+      "epoch": 1.436055469953775,
+      "grad_norm": 2.377561092376709,
+      "learning_rate": 2.925390178911306e-06,
+      "loss": 1.372,
+      "step": 1398
+    },
+    {
+      "epoch": 1.4370826913199795,
+      "grad_norm": 2.5543324947357178,
+      "learning_rate": 2.9234868671488393e-06,
+      "loss": 1.3846,
+      "step": 1399
+    },
+    {
+      "epoch": 1.4381099126861838,
+      "grad_norm": 2.6063973903656006,
+      "learning_rate": 2.9215835553863727e-06,
+      "loss": 1.3989,
+      "step": 1400
+    },
+    {
+      "epoch": 1.4381099126861838,
+      "eval_loss": 1.7730437517166138,
+      "eval_runtime": 38.0407,
+      "eval_samples_per_second": 26.288,
+      "eval_steps_per_second": 1.104,
+      "step": 1400
+    },
+    {
+      "epoch": 1.4391371340523884,
+      "grad_norm": 2.4241244792938232,
+      "learning_rate": 2.9196802436239056e-06,
+      "loss": 1.3945,
+      "step": 1401
+    },
+    {
+      "epoch": 1.4401643554185928,
+      "grad_norm": 2.306528329849243,
+      "learning_rate": 2.917776931861439e-06,
+      "loss": 1.3686,
+      "step": 1402
+    },
+    {
+      "epoch": 1.4411915767847971,
+      "grad_norm": 2.4980850219726562,
+      "learning_rate": 2.9158736200989723e-06,
+      "loss": 1.3714,
+      "step": 1403
+    },
+    {
+      "epoch": 1.4422187981510015,
+      "grad_norm": 2.385195732116699,
+      "learning_rate": 2.9139703083365057e-06,
+      "loss": 1.3706,
+      "step": 1404
+    },
+    {
+      "epoch": 1.4432460195172059,
+      "grad_norm": 2.4087095260620117,
+      "learning_rate": 2.912066996574039e-06,
+      "loss": 1.3967,
+      "step": 1405
+    },
+    {
+      "epoch": 1.4442732408834105,
+      "grad_norm": 2.470900535583496,
+      "learning_rate": 2.910163684811572e-06,
+      "loss": 1.3351,
+      "step": 1406
+    },
+    {
+      "epoch": 1.4453004622496148,
+      "grad_norm": 2.8330743312835693,
+      "learning_rate": 2.9082603730491054e-06,
+      "loss": 1.4413,
+      "step": 1407
+    },
+    {
+      "epoch": 1.4463276836158192,
+      "grad_norm": 2.5553359985351562,
+      "learning_rate": 2.9063570612866387e-06,
+      "loss": 1.4207,
+      "step": 1408
+    },
+    {
+      "epoch": 1.4473549049820236,
+      "grad_norm": 2.509373903274536,
+      "learning_rate": 2.904453749524172e-06,
+      "loss": 1.3387,
+      "step": 1409
+    },
+    {
+      "epoch": 1.448382126348228,
+      "grad_norm": 2.5485401153564453,
+      "learning_rate": 2.902550437761706e-06,
+      "loss": 1.3765,
+      "step": 1410
+    },
+    {
+      "epoch": 1.448382126348228,
+      "eval_loss": 1.7703789472579956,
+      "eval_runtime": 37.7078,
+      "eval_samples_per_second": 26.52,
+      "eval_steps_per_second": 1.114,
+      "step": 1410
+    },
+    {
+      "epoch": 1.4494093477144325,
+      "grad_norm": 2.502002716064453,
+      "learning_rate": 2.9006471259992392e-06,
+      "loss": 1.3842,
+      "step": 1411
+    },
+    {
+      "epoch": 1.450436569080637,
+      "grad_norm": 2.4621520042419434,
+      "learning_rate": 2.8987438142367726e-06,
+      "loss": 1.3809,
+      "step": 1412
+    },
+    {
+      "epoch": 1.4514637904468413,
+      "grad_norm": 2.4255263805389404,
+      "learning_rate": 2.8968405024743055e-06,
+      "loss": 1.4411,
+      "step": 1413
+    },
+    {
+      "epoch": 1.4524910118130456,
+      "grad_norm": 2.2450802326202393,
+      "learning_rate": 2.894937190711839e-06,
+      "loss": 1.3389,
+      "step": 1414
+    },
+    {
+      "epoch": 1.45351823317925,
+      "grad_norm": 2.502903938293457,
+      "learning_rate": 2.8930338789493723e-06,
+      "loss": 1.4066,
+      "step": 1415
+    },
+    {
+      "epoch": 1.4545454545454546,
+      "grad_norm": 2.3685200214385986,
+      "learning_rate": 2.8911305671869056e-06,
+      "loss": 1.3995,
+      "step": 1416
+    },
+    {
+      "epoch": 1.455572675911659,
+      "grad_norm": 2.355095624923706,
+      "learning_rate": 2.889227255424439e-06,
+      "loss": 1.3352,
+      "step": 1417
+    },
+    {
+      "epoch": 1.4565998972778633,
+      "grad_norm": 2.363035202026367,
+      "learning_rate": 2.887323943661972e-06,
+      "loss": 1.3792,
+      "step": 1418
+    },
+    {
+      "epoch": 1.457627118644068,
+      "grad_norm": 2.451078176498413,
+      "learning_rate": 2.8854206318995053e-06,
+      "loss": 1.358,
+      "step": 1419
+    },
+    {
+      "epoch": 1.4586543400102723,
+      "grad_norm": 2.554501533508301,
+      "learning_rate": 2.8835173201370386e-06,
+      "loss": 1.3811,
+      "step": 1420
+    },
+    {
+      "epoch": 1.4586543400102723,
+      "eval_loss": 1.773111343383789,
+      "eval_runtime": 37.7693,
+      "eval_samples_per_second": 26.477,
+      "eval_steps_per_second": 1.112,
+      "step": 1420
+    },
+    {
+      "epoch": 1.4596815613764766,
+      "grad_norm": 2.556537389755249,
+      "learning_rate": 2.881614008374572e-06,
+      "loss": 1.4397,
+      "step": 1421
+    },
+    {
+      "epoch": 1.460708782742681,
+      "grad_norm": 2.5945091247558594,
+      "learning_rate": 2.8797106966121054e-06,
+      "loss": 1.3976,
+      "step": 1422
+    },
+    {
+      "epoch": 1.4617360041088854,
+      "grad_norm": 2.5158495903015137,
+      "learning_rate": 2.8778073848496383e-06,
+      "loss": 1.381,
+      "step": 1423
+    },
+    {
+      "epoch": 1.46276322547509,
+      "grad_norm": 2.359591007232666,
+      "learning_rate": 2.8759040730871717e-06,
+      "loss": 1.4156,
+      "step": 1424
+    },
+    {
+      "epoch": 1.4637904468412943,
+      "grad_norm": 2.3388748168945312,
+      "learning_rate": 2.874000761324705e-06,
+      "loss": 1.3808,
+      "step": 1425
+    },
+    {
+      "epoch": 1.4648176682074987,
+      "grad_norm": 2.438465118408203,
+      "learning_rate": 2.8720974495622384e-06,
+      "loss": 1.4158,
+      "step": 1426
+    },
+    {
+      "epoch": 1.465844889573703,
+      "grad_norm": 2.5624589920043945,
+      "learning_rate": 2.8701941377997718e-06,
+      "loss": 1.3917,
+      "step": 1427
+    },
+    {
+      "epoch": 1.4668721109399074,
+      "grad_norm": 2.467921495437622,
+      "learning_rate": 2.8682908260373047e-06,
+      "loss": 1.4316,
+      "step": 1428
+    },
+    {
+      "epoch": 1.467899332306112,
+      "grad_norm": 2.6034083366394043,
+      "learning_rate": 2.866387514274838e-06,
+      "loss": 1.4058,
+      "step": 1429
+    },
+    {
+      "epoch": 1.4689265536723164,
+      "grad_norm": 2.560844898223877,
+      "learning_rate": 2.864484202512372e-06,
+      "loss": 1.4114,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4689265536723164,
+      "eval_loss": 1.7799324989318848,
+      "eval_runtime": 37.8814,
+      "eval_samples_per_second": 26.398,
+      "eval_steps_per_second": 1.109,
+      "step": 1430
+    },
+    {
+      "epoch": 1.4699537750385208,
+      "grad_norm": 2.469799280166626,
+      "learning_rate": 2.8625808907499052e-06,
+      "loss": 1.3767,
+      "step": 1431
+    },
+    {
+      "epoch": 1.4709809964047253,
+      "grad_norm": 2.408665895462036,
+      "learning_rate": 2.8606775789874386e-06,
+      "loss": 1.4064,
+      "step": 1432
+    },
+    {
+      "epoch": 1.4720082177709297,
+      "grad_norm": 2.497950553894043,
+      "learning_rate": 2.858774267224972e-06,
+      "loss": 1.3747,
+      "step": 1433
+    },
+    {
+      "epoch": 1.473035439137134,
+      "grad_norm": 2.6274573802948,
+      "learning_rate": 2.8568709554625053e-06,
+      "loss": 1.3994,
+      "step": 1434
+    },
+    {
+      "epoch": 1.4740626605033385,
+      "grad_norm": 2.545614719390869,
+      "learning_rate": 2.8549676437000382e-06,
+      "loss": 1.4354,
+      "step": 1435
+    },
+    {
+      "epoch": 1.4750898818695428,
+      "grad_norm": 2.796013355255127,
+      "learning_rate": 2.8530643319375716e-06,
+      "loss": 1.4081,
+      "step": 1436
+    },
+    {
+      "epoch": 1.4761171032357474,
+      "grad_norm": 2.68571400642395,
+      "learning_rate": 2.851161020175105e-06,
+      "loss": 1.3833,
+      "step": 1437
+    },
+    {
+      "epoch": 1.4771443246019518,
+      "grad_norm": 2.455244541168213,
+      "learning_rate": 2.8492577084126383e-06,
+      "loss": 1.3587,
+      "step": 1438
+    },
+    {
+      "epoch": 1.4781715459681561,
+      "grad_norm": 2.552747964859009,
+      "learning_rate": 2.8473543966501717e-06,
+      "loss": 1.4004,
+      "step": 1439
+    },
+    {
+      "epoch": 1.4791987673343605,
+      "grad_norm": 2.9428722858428955,
+      "learning_rate": 2.8454510848877046e-06,
+      "loss": 1.437,
+      "step": 1440
+    },
+    {
+      "epoch": 1.4791987673343605,
+      "eval_loss": 1.7853318452835083,
+      "eval_runtime": 37.8848,
+      "eval_samples_per_second": 26.396,
+      "eval_steps_per_second": 1.109,
+      "step": 1440
+    },
+    {
+      "epoch": 1.4802259887005649,
+      "grad_norm": 2.686199188232422,
+      "learning_rate": 2.843547773125238e-06,
+      "loss": 1.3901,
+      "step": 1441
+    },
+    {
+      "epoch": 1.4812532100667695,
+      "grad_norm": 2.47476863861084,
+      "learning_rate": 2.8416444613627714e-06,
+      "loss": 1.3887,
+      "step": 1442
+    },
+    {
+      "epoch": 1.4822804314329738,
+      "grad_norm": 2.425445795059204,
+      "learning_rate": 2.8397411496003047e-06,
+      "loss": 1.3502,
+      "step": 1443
+    },
+    {
+      "epoch": 1.4833076527991782,
+      "grad_norm": 2.529341459274292,
+      "learning_rate": 2.837837837837838e-06,
+      "loss": 1.3973,
+      "step": 1444
+    },
+    {
+      "epoch": 1.4843348741653826,
+      "grad_norm": 2.3861210346221924,
+      "learning_rate": 2.835934526075371e-06,
+      "loss": 1.4072,
+      "step": 1445
+    },
+    {
+      "epoch": 1.485362095531587,
+      "grad_norm": 2.359403371810913,
+      "learning_rate": 2.8340312143129044e-06,
+      "loss": 1.3553,
+      "step": 1446
+    },
+    {
+      "epoch": 1.4863893168977915,
+      "grad_norm": 2.417792558670044,
+      "learning_rate": 2.8321279025504377e-06,
+      "loss": 1.3539,
+      "step": 1447
+    },
+    {
+      "epoch": 1.487416538263996,
+      "grad_norm": 2.7177648544311523,
+      "learning_rate": 2.830224590787971e-06,
+      "loss": 1.4575,
+      "step": 1448
+    },
+    {
+      "epoch": 1.4884437596302003,
+      "grad_norm": 2.514181613922119,
+      "learning_rate": 2.8283212790255045e-06,
+      "loss": 1.388,
+      "step": 1449
+    },
+    {
+      "epoch": 1.4894709809964048,
+      "grad_norm": 2.4293761253356934,
+      "learning_rate": 2.8264179672630383e-06,
+      "loss": 1.4143,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4894709809964048,
+      "eval_loss": 1.781496286392212,
+      "eval_runtime": 37.8942,
+      "eval_samples_per_second": 26.389,
+      "eval_steps_per_second": 1.108,
+      "step": 1450
+    },
+    {
+      "epoch": 1.4904982023626092,
+      "grad_norm": 2.495591878890991,
+      "learning_rate": 2.8245146555005716e-06,
+      "loss": 1.3809,
+      "step": 1451
+    },
+    {
+      "epoch": 1.4915254237288136,
+      "grad_norm": 2.520158290863037,
+      "learning_rate": 2.8226113437381046e-06,
+      "loss": 1.3297,
+      "step": 1452
+    },
+    {
+      "epoch": 1.492552645095018,
+      "grad_norm": 2.403061866760254,
+      "learning_rate": 2.820708031975638e-06,
+      "loss": 1.3985,
+      "step": 1453
+    },
+    {
+      "epoch": 1.4935798664612223,
+      "grad_norm": 2.606884479522705,
+      "learning_rate": 2.8188047202131713e-06,
+      "loss": 1.4179,
+      "step": 1454
+    },
+    {
+      "epoch": 1.494607087827427,
+      "grad_norm": 2.4543404579162598,
+      "learning_rate": 2.8169014084507046e-06,
+      "loss": 1.4325,
+      "step": 1455
+    },
+    {
+      "epoch": 1.4956343091936313,
+      "grad_norm": 2.3947877883911133,
+      "learning_rate": 2.814998096688238e-06,
+      "loss": 1.3876,
+      "step": 1456
+    },
+    {
+      "epoch": 1.4966615305598356,
+      "grad_norm": 2.4632408618927,
+      "learning_rate": 2.813094784925771e-06,
+      "loss": 1.3392,
+      "step": 1457
+    },
+    {
+      "epoch": 1.49768875192604,
+      "grad_norm": 2.507312059402466,
+      "learning_rate": 2.8111914731633043e-06,
+      "loss": 1.4351,
+      "step": 1458
+    },
+    {
+      "epoch": 1.4987159732922444,
+      "grad_norm": 2.6338255405426025,
+      "learning_rate": 2.8092881614008377e-06,
+      "loss": 1.389,
+      "step": 1459
+    },
+    {
+      "epoch": 1.499743194658449,
+      "grad_norm": 2.3906662464141846,
+      "learning_rate": 2.807384849638371e-06,
+      "loss": 1.3655,
+      "step": 1460
+    },
+    {
+      "epoch": 1.499743194658449,
+      "eval_loss": 1.7716007232666016,
+      "eval_runtime": 37.8931,
+      "eval_samples_per_second": 26.39,
+      "eval_steps_per_second": 1.108,
+      "step": 1460
+    },
+    {
+      "epoch": 1.5007704160246533,
+      "grad_norm": 2.4841926097869873,
+      "learning_rate": 2.8054815378759044e-06,
+      "loss": 1.4046,
+      "step": 1461
+    },
+    {
+      "epoch": 1.5017976373908577,
+      "grad_norm": 2.4934608936309814,
+      "learning_rate": 2.8035782261134378e-06,
+      "loss": 1.3781,
+      "step": 1462
+    },
+    {
+      "epoch": 1.5028248587570623,
+      "grad_norm": 2.538504123687744,
+      "learning_rate": 2.8016749143509707e-06,
+      "loss": 1.383,
+      "step": 1463
+    },
+    {
+      "epoch": 1.5038520801232664,
+      "grad_norm": 2.461595058441162,
+      "learning_rate": 2.799771602588504e-06,
+      "loss": 1.3892,
+      "step": 1464
+    },
+    {
+      "epoch": 1.504879301489471,
+      "grad_norm": 2.3686397075653076,
+      "learning_rate": 2.7978682908260374e-06,
+      "loss": 1.4037,
+      "step": 1465
+    },
+    {
+      "epoch": 1.5059065228556754,
+      "grad_norm": 2.47755765914917,
+      "learning_rate": 2.795964979063571e-06,
+      "loss": 1.4173,
+      "step": 1466
+    },
+    {
+      "epoch": 1.5069337442218798,
+      "grad_norm": 2.504168748855591,
+      "learning_rate": 2.794061667301104e-06,
+      "loss": 1.3955,
+      "step": 1467
+    },
+    {
+      "epoch": 1.5079609655880843,
+      "grad_norm": 2.4311411380767822,
+      "learning_rate": 2.792158355538637e-06,
+      "loss": 1.4079,
+      "step": 1468
+    },
+    {
+      "epoch": 1.5089881869542885,
+      "grad_norm": 2.523200273513794,
+      "learning_rate": 2.7902550437761705e-06,
+      "loss": 1.4019,
+      "step": 1469
+    },
+    {
+      "epoch": 1.510015408320493,
+      "grad_norm": 2.3743972778320312,
+      "learning_rate": 2.7883517320137042e-06,
+      "loss": 1.3677,
+      "step": 1470
+    },
+    {
+      "epoch": 1.510015408320493,
+      "eval_loss": 1.7622252702713013,
+      "eval_runtime": 37.8933,
+      "eval_samples_per_second": 26.39,
+      "eval_steps_per_second": 1.108,
+      "step": 1470
+    },
+    {
+      "epoch": 1.5110426296866974,
+      "grad_norm": 2.4073667526245117,
+      "learning_rate": 2.7864484202512376e-06,
+      "loss": 1.3623,
+      "step": 1471
+    },
+    {
+      "epoch": 1.5120698510529018,
+      "grad_norm": 2.5012941360473633,
+      "learning_rate": 2.784545108488771e-06,
+      "loss": 1.382,
+      "step": 1472
+    },
+    {
+      "epoch": 1.5130970724191064,
+      "grad_norm": 2.425480365753174,
+      "learning_rate": 2.7826417967263043e-06,
+      "loss": 1.3896,
+      "step": 1473
+    },
+    {
+      "epoch": 1.5141242937853108,
+      "grad_norm": 2.4600207805633545,
+      "learning_rate": 2.7807384849638373e-06,
+      "loss": 1.4364,
+      "step": 1474
+    },
+    {
+      "epoch": 1.5151515151515151,
+      "grad_norm": 2.5489892959594727,
+      "learning_rate": 2.7788351732013706e-06,
+      "loss": 1.3832,
+      "step": 1475
+    },
+    {
+      "epoch": 1.5161787365177197,
+      "grad_norm": 2.4228315353393555,
+      "learning_rate": 2.776931861438904e-06,
+      "loss": 1.3755,
+      "step": 1476
+    },
+    {
+      "epoch": 1.5172059578839239,
+      "grad_norm": 2.519340991973877,
+      "learning_rate": 2.7750285496764374e-06,
+      "loss": 1.3915,
+      "step": 1477
+    },
+    {
+      "epoch": 1.5182331792501285,
+      "grad_norm": 2.3544020652770996,
+      "learning_rate": 2.7731252379139707e-06,
+      "loss": 1.394,
+      "step": 1478
+    },
+    {
+      "epoch": 1.5192604006163328,
+      "grad_norm": 2.4037892818450928,
+      "learning_rate": 2.771221926151504e-06,
+      "loss": 1.3396,
+      "step": 1479
+    },
+    {
+      "epoch": 1.5202876219825372,
+      "grad_norm": 2.409222364425659,
+      "learning_rate": 2.769318614389037e-06,
+      "loss": 1.4052,
+      "step": 1480
+    },
+    {
+      "epoch": 1.5202876219825372,
+      "eval_loss": 1.7622238397598267,
+      "eval_runtime": 37.9064,
+      "eval_samples_per_second": 26.381,
+      "eval_steps_per_second": 1.108,
+      "step": 1480
+    },
+    {
+      "epoch": 1.5213148433487418,
+      "grad_norm": 2.3649749755859375,
+      "learning_rate": 2.7674153026265704e-06,
+      "loss": 1.401,
+      "step": 1481
+    },
+    {
+      "epoch": 1.522342064714946,
+      "grad_norm": 2.4484262466430664,
+      "learning_rate": 2.7655119908641037e-06,
+      "loss": 1.4026,
+      "step": 1482
+    },
+    {
+      "epoch": 1.5233692860811505,
+      "grad_norm": 2.492107629776001,
+      "learning_rate": 2.763608679101637e-06,
+      "loss": 1.3985,
+      "step": 1483
+    },
+    {
+      "epoch": 1.5243965074473549,
+      "grad_norm": 2.5192503929138184,
+      "learning_rate": 2.7617053673391705e-06,
+      "loss": 1.4282,
+      "step": 1484
+    },
+    {
+      "epoch": 1.5254237288135593,
+      "grad_norm": 2.471250534057617,
+      "learning_rate": 2.7598020555767034e-06,
+      "loss": 1.4217,
+      "step": 1485
+    },
+    {
+      "epoch": 1.5264509501797638,
+      "grad_norm": 2.2985496520996094,
+      "learning_rate": 2.7578987438142368e-06,
+      "loss": 1.3599,
+      "step": 1486
+    },
+    {
+      "epoch": 1.5274781715459682,
+      "grad_norm": 2.492047071456909,
+      "learning_rate": 2.75599543205177e-06,
+      "loss": 1.4036,
+      "step": 1487
+    },
+    {
+      "epoch": 1.5285053929121726,
+      "grad_norm": 2.5131945610046387,
+      "learning_rate": 2.7540921202893035e-06,
+      "loss": 1.4206,
+      "step": 1488
+    },
+    {
+      "epoch": 1.529532614278377,
+      "grad_norm": 2.619874954223633,
+      "learning_rate": 2.752188808526837e-06,
+      "loss": 1.4055,
+      "step": 1489
+    },
+    {
+      "epoch": 1.5305598356445813,
+      "grad_norm": 2.4526877403259277,
+      "learning_rate": 2.75028549676437e-06,
+      "loss": 1.3955,
+      "step": 1490
+    },
+    {
+      "epoch": 1.5305598356445813,
+      "eval_loss": 1.758982539176941,
+      "eval_runtime": 37.8756,
+      "eval_samples_per_second": 26.402,
+      "eval_steps_per_second": 1.109,
+      "step": 1490
+    },
+    {
+      "epoch": 1.531587057010786,
+      "grad_norm": 2.4925529956817627,
+      "learning_rate": 2.7483821850019036e-06,
+      "loss": 1.3899,
+      "step": 1491
+    },
+    {
+      "epoch": 1.5326142783769903,
+      "grad_norm": 2.3789961338043213,
+      "learning_rate": 2.746478873239437e-06,
+      "loss": 1.377,
+      "step": 1492
+    },
+    {
+      "epoch": 1.5336414997431946,
+      "grad_norm": 2.4063172340393066,
+      "learning_rate": 2.7445755614769703e-06,
+      "loss": 1.384,
+      "step": 1493
+    },
+    {
+      "epoch": 1.5346687211093992,
+      "grad_norm": 2.3189985752105713,
+      "learning_rate": 2.7426722497145037e-06,
+      "loss": 1.3635,
+      "step": 1494
+    },
+    {
+      "epoch": 1.5356959424756034,
+      "grad_norm": 2.4792535305023193,
+      "learning_rate": 2.740768937952037e-06,
+      "loss": 1.3676,
+      "step": 1495
+    },
+    {
+      "epoch": 1.536723163841808,
+      "grad_norm": 2.342355489730835,
+      "learning_rate": 2.7388656261895704e-06,
+      "loss": 1.3812,
+      "step": 1496
+    },
+    {
+      "epoch": 1.5377503852080123,
+      "grad_norm": 2.556312084197998,
+      "learning_rate": 2.7369623144271033e-06,
+      "loss": 1.4529,
+      "step": 1497
+    },
+    {
+      "epoch": 1.5387776065742167,
+      "grad_norm": 2.555600643157959,
+      "learning_rate": 2.7350590026646367e-06,
+      "loss": 1.3819,
+      "step": 1498
+    },
+    {
+      "epoch": 1.5398048279404213,
+      "grad_norm": 2.4179797172546387,
+      "learning_rate": 2.73315569090217e-06,
+      "loss": 1.3383,
+      "step": 1499
+    },
+    {
+      "epoch": 1.5408320493066254,
+      "grad_norm": 2.5075631141662598,
+      "learning_rate": 2.7312523791397034e-06,
+      "loss": 1.4217,
+      "step": 1500
+    },
+    {
+      "epoch": 1.5408320493066254,
+      "eval_loss": 1.769559621810913,
+      "eval_runtime": 37.8701,
+      "eval_samples_per_second": 26.406,
+      "eval_steps_per_second": 1.109,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2919,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "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": 4.67596411404288e+17,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}