Training in progress, step 189, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -20,13 +20,13 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": "unsloth",
|
| 22 |
"target_modules": [
|
| 23 |
-
"q_proj",
|
| 24 |
"k_proj",
|
| 25 |
-
"
|
| 26 |
-
"gate_proj",
|
| 27 |
"v_proj",
|
|
|
|
|
|
|
| 28 |
"down_proj",
|
| 29 |
-
"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": "unsloth",
|
| 22 |
"target_modules": [
|
|
|
|
| 23 |
"k_proj",
|
| 24 |
+
"q_proj",
|
|
|
|
| 25 |
"v_proj",
|
| 26 |
+
"o_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
"down_proj",
|
| 29 |
+
"gate_proj"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 60010048
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e625bcbda58a2077a2f53862e647a4496991959b8e386df7a79f732236f18e9f
|
| 3 |
size 60010048
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 30427860
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03e4632d18d4e91d695ae7e07e8db236469a783bace61d269def1460bacaf75a
|
| 3 |
size 30427860
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96200d4ca8dc27a2feac013eb62dd34ec99b5598bcdd66554ba3bfa0cfc0128f
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5aa8b796ae61fd6c2a8cd520f49549ae764a41fd16083782a810533cbf63c112
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,142 +1,1339 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 1
|
| 17 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
{
|
| 19 |
"epoch": 0.0043895747599451305,
|
| 20 |
-
"grad_norm": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
"learning_rate": 1.0000000000000002e-06,
|
| 22 |
-
"loss": 2.
|
| 23 |
-
"step":
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.006584362139917695,
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss": 2.
|
| 30 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.008779149519890261,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 2.
|
| 37 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.010973936899862825,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss": 2.
|
| 44 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.01316872427983539,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.015363511659807956,
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 2.
|
| 58 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.017558299039780522,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 2.
|
| 65 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.019753086419753086,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 2.
|
| 72 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.02194787379972565,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 2.
|
| 79 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.024142661179698217,
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 2.
|
| 86 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.02633744855967078,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 2.
|
| 93 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.02853223593964335,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss": 2.
|
| 100 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.030727023319615913,
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 2.
|
| 107 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.03292181069958848,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 2.
|
| 114 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.035116598079561044,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss":
|
| 121 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.03731138545953361,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 2.
|
| 128 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.03950617283950617,
|
| 132 |
-
"grad_norm": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 133 |
"learning_rate": 9e-06,
|
| 134 |
-
"loss":
|
| 135 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 136 |
}
|
| 137 |
],
|
| 138 |
"logging_steps": 1,
|
| 139 |
-
"max_steps":
|
| 140 |
"num_input_tokens_seen": 0,
|
| 141 |
"num_train_epochs": 1,
|
| 142 |
"save_steps": 1,
|
|
@@ -152,7 +1349,7 @@
|
|
| 152 |
"attributes": {}
|
| 153 |
}
|
| 154 |
},
|
| 155 |
-
"total_flos":
|
| 156 |
"train_batch_size": 4,
|
| 157 |
"trial_name": null,
|
| 158 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.2074074074074074,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 189,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0010973936899862826,
|
| 13 |
+
"grad_norm": 0.5590636134147644,
|
| 14 |
+
"learning_rate": 2.0000000000000002e-07,
|
| 15 |
+
"loss": 2.0815,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0021947873799725653,
|
| 20 |
+
"grad_norm": 0.5765010714530945,
|
| 21 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 22 |
+
"loss": 2.1074,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0032921810699588477,
|
| 27 |
+
"grad_norm": 0.584500253200531,
|
| 28 |
+
"learning_rate": 6.000000000000001e-07,
|
| 29 |
+
"loss": 2.1012,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
{
|
| 33 |
"epoch": 0.0043895747599451305,
|
| 34 |
+
"grad_norm": 0.6246315836906433,
|
| 35 |
+
"learning_rate": 8.000000000000001e-07,
|
| 36 |
+
"loss": 2.1795,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0054869684499314125,
|
| 41 |
+
"grad_norm": 0.5558974146842957,
|
| 42 |
"learning_rate": 1.0000000000000002e-06,
|
| 43 |
+
"loss": 2.0729,
|
| 44 |
+
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.006584362139917695,
|
| 48 |
+
"grad_norm": 0.6098957657814026,
|
| 49 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 50 |
+
"loss": 2.1216,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.007681755829903978,
|
| 55 |
+
"grad_norm": 0.5948249697685242,
|
| 56 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 57 |
+
"loss": 2.1381,
|
| 58 |
+
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.008779149519890261,
|
| 62 |
+
"grad_norm": 0.6547859311103821,
|
| 63 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 64 |
+
"loss": 2.2398,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.009876543209876543,
|
| 69 |
+
"grad_norm": 0.6762146353721619,
|
| 70 |
+
"learning_rate": 1.8000000000000001e-06,
|
| 71 |
+
"loss": 2.1872,
|
| 72 |
+
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.010973936899862825,
|
| 76 |
+
"grad_norm": 0.6071422100067139,
|
| 77 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 78 |
+
"loss": 2.1451,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.012071330589849109,
|
| 83 |
+
"grad_norm": 0.5874962210655212,
|
| 84 |
+
"learning_rate": 2.2e-06,
|
| 85 |
+
"loss": 2.1632,
|
| 86 |
+
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.01316872427983539,
|
| 90 |
+
"grad_norm": 0.5862544775009155,
|
| 91 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 92 |
+
"loss": 2.1152,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.014266117969821674,
|
| 97 |
+
"grad_norm": 0.614499032497406,
|
| 98 |
+
"learning_rate": 2.6e-06,
|
| 99 |
+
"loss": 2.1906,
|
| 100 |
+
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.015363511659807956,
|
| 104 |
+
"grad_norm": 0.6741944551467896,
|
| 105 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 106 |
+
"loss": 2.2436,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.01646090534979424,
|
| 111 |
+
"grad_norm": 0.5827784538269043,
|
| 112 |
+
"learning_rate": 3e-06,
|
| 113 |
+
"loss": 2.0787,
|
| 114 |
+
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.017558299039780522,
|
| 118 |
+
"grad_norm": 0.5573399066925049,
|
| 119 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 120 |
+
"loss": 2.0608,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.018655692729766804,
|
| 125 |
+
"grad_norm": 0.5720248818397522,
|
| 126 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 127 |
+
"loss": 2.1152,
|
| 128 |
+
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.019753086419753086,
|
| 132 |
+
"grad_norm": 0.5938870310783386,
|
| 133 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 134 |
+
"loss": 2.1217,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.020850480109739368,
|
| 139 |
+
"grad_norm": 0.6471643447875977,
|
| 140 |
+
"learning_rate": 3.8000000000000005e-06,
|
| 141 |
+
"loss": 2.1649,
|
| 142 |
+
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.02194787379972565,
|
| 146 |
+
"grad_norm": 0.6734815835952759,
|
| 147 |
+
"learning_rate": 4.000000000000001e-06,
|
| 148 |
+
"loss": 2.1458,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.023045267489711935,
|
| 153 |
+
"grad_norm": 0.5911048054695129,
|
| 154 |
+
"learning_rate": 4.2000000000000004e-06,
|
| 155 |
+
"loss": 2.0573,
|
| 156 |
+
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.024142661179698217,
|
| 160 |
+
"grad_norm": 0.5842518210411072,
|
| 161 |
+
"learning_rate": 4.4e-06,
|
| 162 |
+
"loss": 2.0675,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0252400548696845,
|
| 167 |
+
"grad_norm": 0.6548774242401123,
|
| 168 |
+
"learning_rate": 4.600000000000001e-06,
|
| 169 |
+
"loss": 2.1,
|
| 170 |
+
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.02633744855967078,
|
| 174 |
+
"grad_norm": 0.5978567600250244,
|
| 175 |
+
"learning_rate": 4.800000000000001e-06,
|
| 176 |
+
"loss": 2.1234,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.027434842249657063,
|
| 181 |
+
"grad_norm": 0.6166532635688782,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 2.1784,
|
| 184 |
+
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.02853223593964335,
|
| 188 |
+
"grad_norm": 0.5846672058105469,
|
| 189 |
+
"learning_rate": 5.2e-06,
|
| 190 |
+
"loss": 2.1453,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.02962962962962963,
|
| 195 |
+
"grad_norm": 0.5325105786323547,
|
| 196 |
+
"learning_rate": 5.400000000000001e-06,
|
| 197 |
+
"loss": 2.0348,
|
| 198 |
+
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.030727023319615913,
|
| 202 |
+
"grad_norm": 0.5196258425712585,
|
| 203 |
+
"learning_rate": 5.600000000000001e-06,
|
| 204 |
+
"loss": 2.0484,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.031824417009602195,
|
| 209 |
+
"grad_norm": 0.5424355268478394,
|
| 210 |
+
"learning_rate": 5.8e-06,
|
| 211 |
+
"loss": 2.0728,
|
| 212 |
+
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.03292181069958848,
|
| 216 |
+
"grad_norm": 0.5482433438301086,
|
| 217 |
+
"learning_rate": 6e-06,
|
| 218 |
+
"loss": 2.051,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.03401920438957476,
|
| 223 |
+
"grad_norm": 0.5127721428871155,
|
| 224 |
+
"learning_rate": 6.200000000000001e-06,
|
| 225 |
+
"loss": 2.0508,
|
| 226 |
+
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.035116598079561044,
|
| 230 |
+
"grad_norm": 0.5115760564804077,
|
| 231 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 232 |
+
"loss": 1.9822,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.03621399176954732,
|
| 237 |
+
"grad_norm": 0.5571993589401245,
|
| 238 |
+
"learning_rate": 6.600000000000001e-06,
|
| 239 |
+
"loss": 2.0929,
|
| 240 |
+
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.03731138545953361,
|
| 244 |
+
"grad_norm": 0.5175538063049316,
|
| 245 |
+
"learning_rate": 6.800000000000001e-06,
|
| 246 |
+
"loss": 2.0776,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.038408779149519894,
|
| 251 |
+
"grad_norm": 0.5197016596794128,
|
| 252 |
+
"learning_rate": 7e-06,
|
| 253 |
+
"loss": 2.0747,
|
| 254 |
+
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.03950617283950617,
|
| 258 |
+
"grad_norm": 0.5387422442436218,
|
| 259 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 260 |
+
"loss": 2.0864,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.04060356652949246,
|
| 265 |
+
"grad_norm": 0.5122085809707642,
|
| 266 |
+
"learning_rate": 7.4e-06,
|
| 267 |
+
"loss": 2.0737,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.041700960219478736,
|
| 272 |
+
"grad_norm": 0.6023052334785461,
|
| 273 |
+
"learning_rate": 7.600000000000001e-06,
|
| 274 |
+
"loss": 2.0519,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.04279835390946502,
|
| 279 |
+
"grad_norm": 0.5678107738494873,
|
| 280 |
+
"learning_rate": 7.800000000000002e-06,
|
| 281 |
+
"loss": 2.1175,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.0438957475994513,
|
| 286 |
+
"grad_norm": 0.5373347401618958,
|
| 287 |
+
"learning_rate": 8.000000000000001e-06,
|
| 288 |
+
"loss": 2.0667,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.044993141289437585,
|
| 293 |
+
"grad_norm": 0.5342459082603455,
|
| 294 |
+
"learning_rate": 8.2e-06,
|
| 295 |
+
"loss": 2.081,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.04609053497942387,
|
| 300 |
+
"grad_norm": 0.6377962231636047,
|
| 301 |
+
"learning_rate": 8.400000000000001e-06,
|
| 302 |
+
"loss": 2.1576,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.04718792866941015,
|
| 307 |
+
"grad_norm": 0.46719589829444885,
|
| 308 |
+
"learning_rate": 8.6e-06,
|
| 309 |
+
"loss": 1.9755,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.048285322359396435,
|
| 314 |
+
"grad_norm": 0.4903099834918976,
|
| 315 |
+
"learning_rate": 8.8e-06,
|
| 316 |
+
"loss": 2.0195,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.04938271604938271,
|
| 321 |
+
"grad_norm": 0.460610955953598,
|
| 322 |
"learning_rate": 9e-06,
|
| 323 |
+
"loss": 1.9456,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.050480109739369,
|
| 328 |
+
"grad_norm": 0.45545804500579834,
|
| 329 |
+
"learning_rate": 9.200000000000002e-06,
|
| 330 |
+
"loss": 2.0255,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.051577503429355284,
|
| 335 |
+
"grad_norm": 0.4546717405319214,
|
| 336 |
+
"learning_rate": 9.4e-06,
|
| 337 |
+
"loss": 1.9741,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.05267489711934156,
|
| 342 |
+
"grad_norm": 0.46261245012283325,
|
| 343 |
+
"learning_rate": 9.600000000000001e-06,
|
| 344 |
+
"loss": 2.0104,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.05377229080932785,
|
| 349 |
+
"grad_norm": 0.43465015292167664,
|
| 350 |
+
"learning_rate": 9.800000000000001e-06,
|
| 351 |
+
"loss": 1.9356,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.05486968449931413,
|
| 356 |
+
"grad_norm": 0.42450013756752014,
|
| 357 |
+
"learning_rate": 1e-05,
|
| 358 |
+
"loss": 1.9312,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.05596707818930041,
|
| 363 |
+
"grad_norm": 0.4223531484603882,
|
| 364 |
+
"learning_rate": 1.02e-05,
|
| 365 |
+
"loss": 1.9022,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.0570644718792867,
|
| 370 |
+
"grad_norm": 0.39140722155570984,
|
| 371 |
+
"learning_rate": 1.04e-05,
|
| 372 |
+
"loss": 1.9784,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.058161865569272976,
|
| 377 |
+
"grad_norm": 0.4256257712841034,
|
| 378 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 379 |
+
"loss": 1.9278,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.05925925925925926,
|
| 384 |
+
"grad_norm": 0.45769235491752625,
|
| 385 |
+
"learning_rate": 1.0800000000000002e-05,
|
| 386 |
+
"loss": 2.031,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.06035665294924554,
|
| 391 |
+
"grad_norm": 0.4826626181602478,
|
| 392 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 393 |
+
"loss": 1.9647,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.061454046639231825,
|
| 398 |
+
"grad_norm": 0.4414077699184418,
|
| 399 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 400 |
+
"loss": 2.0007,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.06255144032921811,
|
| 405 |
+
"grad_norm": 0.39386674761772156,
|
| 406 |
+
"learning_rate": 1.14e-05,
|
| 407 |
+
"loss": 1.8676,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.06364883401920439,
|
| 412 |
+
"grad_norm": 0.386751264333725,
|
| 413 |
+
"learning_rate": 1.16e-05,
|
| 414 |
+
"loss": 1.9471,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.06474622770919067,
|
| 419 |
+
"grad_norm": 0.3908196985721588,
|
| 420 |
+
"learning_rate": 1.18e-05,
|
| 421 |
+
"loss": 1.9722,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.06584362139917696,
|
| 426 |
+
"grad_norm": 0.39014488458633423,
|
| 427 |
+
"learning_rate": 1.2e-05,
|
| 428 |
+
"loss": 1.9328,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.06694101508916324,
|
| 433 |
+
"grad_norm": 0.3620125651359558,
|
| 434 |
+
"learning_rate": 1.22e-05,
|
| 435 |
+
"loss": 1.9918,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.06803840877914952,
|
| 440 |
+
"grad_norm": 0.37926608324050903,
|
| 441 |
+
"learning_rate": 1.2400000000000002e-05,
|
| 442 |
+
"loss": 1.8233,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.0691358024691358,
|
| 447 |
+
"grad_norm": 0.44235774874687195,
|
| 448 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 449 |
+
"loss": 1.9558,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.07023319615912209,
|
| 454 |
+
"grad_norm": 0.3922639489173889,
|
| 455 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 456 |
+
"loss": 1.993,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.07133058984910837,
|
| 461 |
+
"grad_norm": 0.4201815128326416,
|
| 462 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 463 |
+
"loss": 1.873,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.07242798353909465,
|
| 468 |
+
"grad_norm": 0.3698742389678955,
|
| 469 |
+
"learning_rate": 1.3200000000000002e-05,
|
| 470 |
+
"loss": 1.8728,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.07352537722908094,
|
| 475 |
+
"grad_norm": 0.38322025537490845,
|
| 476 |
+
"learning_rate": 1.3400000000000002e-05,
|
| 477 |
+
"loss": 1.8322,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.07462277091906722,
|
| 482 |
+
"grad_norm": 0.37198516726493835,
|
| 483 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 484 |
+
"loss": 1.932,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.0757201646090535,
|
| 489 |
+
"grad_norm": 0.36407792568206787,
|
| 490 |
+
"learning_rate": 1.38e-05,
|
| 491 |
+
"loss": 1.826,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.07681755829903979,
|
| 496 |
+
"grad_norm": 0.3603726625442505,
|
| 497 |
+
"learning_rate": 1.4e-05,
|
| 498 |
+
"loss": 1.8001,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.07791495198902607,
|
| 503 |
+
"grad_norm": 0.37135443091392517,
|
| 504 |
+
"learning_rate": 1.4200000000000001e-05,
|
| 505 |
+
"loss": 1.8165,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.07901234567901234,
|
| 510 |
+
"grad_norm": 0.37606292963027954,
|
| 511 |
+
"learning_rate": 1.4400000000000001e-05,
|
| 512 |
+
"loss": 1.7207,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.08010973936899862,
|
| 517 |
+
"grad_norm": 0.3831545114517212,
|
| 518 |
+
"learning_rate": 1.46e-05,
|
| 519 |
+
"loss": 1.8692,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.08120713305898491,
|
| 524 |
+
"grad_norm": 0.3911626935005188,
|
| 525 |
+
"learning_rate": 1.48e-05,
|
| 526 |
+
"loss": 1.8687,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.0823045267489712,
|
| 531 |
+
"grad_norm": 0.39615172147750854,
|
| 532 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 533 |
+
"loss": 1.7774,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.08340192043895747,
|
| 538 |
+
"grad_norm": 0.37161362171173096,
|
| 539 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 540 |
+
"loss": 1.723,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.08449931412894376,
|
| 545 |
+
"grad_norm": 0.3870552182197571,
|
| 546 |
+
"learning_rate": 1.54e-05,
|
| 547 |
+
"loss": 1.7314,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.08559670781893004,
|
| 552 |
+
"grad_norm": 0.4143535792827606,
|
| 553 |
+
"learning_rate": 1.5600000000000003e-05,
|
| 554 |
+
"loss": 1.7912,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.08669410150891632,
|
| 559 |
+
"grad_norm": 0.4270377457141876,
|
| 560 |
+
"learning_rate": 1.58e-05,
|
| 561 |
+
"loss": 1.78,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.0877914951989026,
|
| 566 |
+
"grad_norm": 0.4058167636394501,
|
| 567 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 568 |
+
"loss": 1.7228,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.08888888888888889,
|
| 573 |
+
"grad_norm": 0.42461684346199036,
|
| 574 |
+
"learning_rate": 1.62e-05,
|
| 575 |
+
"loss": 1.8259,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.08998628257887517,
|
| 580 |
+
"grad_norm": 0.37542393803596497,
|
| 581 |
+
"learning_rate": 1.64e-05,
|
| 582 |
+
"loss": 1.7579,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.09108367626886145,
|
| 587 |
+
"grad_norm": 0.4218761622905731,
|
| 588 |
+
"learning_rate": 1.66e-05,
|
| 589 |
+
"loss": 1.8108,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.09218106995884774,
|
| 594 |
+
"grad_norm": 0.4039926528930664,
|
| 595 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 596 |
+
"loss": 1.7625,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.09327846364883402,
|
| 601 |
+
"grad_norm": 0.3840247690677643,
|
| 602 |
+
"learning_rate": 1.7e-05,
|
| 603 |
+
"loss": 1.7455,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.0943758573388203,
|
| 608 |
+
"grad_norm": 0.41403114795684814,
|
| 609 |
+
"learning_rate": 1.72e-05,
|
| 610 |
+
"loss": 1.6384,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.09547325102880659,
|
| 615 |
+
"grad_norm": 0.3860199451446533,
|
| 616 |
+
"learning_rate": 1.7400000000000003e-05,
|
| 617 |
+
"loss": 1.6876,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.09657064471879287,
|
| 622 |
+
"grad_norm": 0.39576929807662964,
|
| 623 |
+
"learning_rate": 1.76e-05,
|
| 624 |
+
"loss": 1.7044,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.09766803840877915,
|
| 629 |
+
"grad_norm": 0.4149666130542755,
|
| 630 |
+
"learning_rate": 1.7800000000000002e-05,
|
| 631 |
+
"loss": 1.657,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.09876543209876543,
|
| 636 |
+
"grad_norm": 0.46741801500320435,
|
| 637 |
+
"learning_rate": 1.8e-05,
|
| 638 |
+
"loss": 1.6955,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.09986282578875172,
|
| 643 |
+
"grad_norm": 0.41203179955482483,
|
| 644 |
+
"learning_rate": 1.8200000000000002e-05,
|
| 645 |
+
"loss": 1.6259,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.100960219478738,
|
| 650 |
+
"grad_norm": 0.4069835841655731,
|
| 651 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 652 |
+
"loss": 1.6547,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.10205761316872428,
|
| 657 |
+
"grad_norm": 0.4733927845954895,
|
| 658 |
+
"learning_rate": 1.86e-05,
|
| 659 |
+
"loss": 1.6419,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.10315500685871057,
|
| 664 |
+
"grad_norm": 0.4243912100791931,
|
| 665 |
+
"learning_rate": 1.88e-05,
|
| 666 |
+
"loss": 1.5943,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.10425240054869685,
|
| 671 |
+
"grad_norm": 0.48016536235809326,
|
| 672 |
+
"learning_rate": 1.9e-05,
|
| 673 |
+
"loss": 1.5303,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.10534979423868313,
|
| 678 |
+
"grad_norm": 0.44309377670288086,
|
| 679 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 680 |
+
"loss": 1.5682,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.1064471879286694,
|
| 685 |
+
"grad_norm": 0.45906150341033936,
|
| 686 |
+
"learning_rate": 1.94e-05,
|
| 687 |
+
"loss": 1.5747,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.1075445816186557,
|
| 692 |
+
"grad_norm": 0.4476592540740967,
|
| 693 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 694 |
+
"loss": 1.5563,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.10864197530864197,
|
| 699 |
+
"grad_norm": 0.432974249124527,
|
| 700 |
+
"learning_rate": 1.98e-05,
|
| 701 |
+
"loss": 1.5782,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.10973936899862825,
|
| 706 |
+
"grad_norm": 0.4596545398235321,
|
| 707 |
+
"learning_rate": 2e-05,
|
| 708 |
+
"loss": 1.5268,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.11083676268861455,
|
| 713 |
+
"grad_norm": 0.49297234416007996,
|
| 714 |
+
"learning_rate": 1.997533908754624e-05,
|
| 715 |
+
"loss": 1.4867,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.11193415637860082,
|
| 720 |
+
"grad_norm": 0.4385370910167694,
|
| 721 |
+
"learning_rate": 1.995067817509248e-05,
|
| 722 |
+
"loss": 1.5288,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.1130315500685871,
|
| 727 |
+
"grad_norm": 0.4075908660888672,
|
| 728 |
+
"learning_rate": 1.992601726263872e-05,
|
| 729 |
+
"loss": 1.5342,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.1141289437585734,
|
| 734 |
+
"grad_norm": 0.4361695647239685,
|
| 735 |
+
"learning_rate": 1.990135635018496e-05,
|
| 736 |
+
"loss": 1.4855,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.11522633744855967,
|
| 741 |
+
"grad_norm": 0.3582554757595062,
|
| 742 |
+
"learning_rate": 1.9876695437731196e-05,
|
| 743 |
+
"loss": 1.5212,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.11632373113854595,
|
| 748 |
+
"grad_norm": 0.39850226044654846,
|
| 749 |
+
"learning_rate": 1.985203452527744e-05,
|
| 750 |
+
"loss": 1.4019,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.11742112482853223,
|
| 755 |
+
"grad_norm": 0.35860705375671387,
|
| 756 |
+
"learning_rate": 1.9827373612823677e-05,
|
| 757 |
+
"loss": 1.401,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.11851851851851852,
|
| 762 |
+
"grad_norm": 0.34796416759490967,
|
| 763 |
+
"learning_rate": 1.9802712700369916e-05,
|
| 764 |
+
"loss": 1.3828,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.1196159122085048,
|
| 769 |
+
"grad_norm": 0.3401014506816864,
|
| 770 |
+
"learning_rate": 1.9778051787916155e-05,
|
| 771 |
+
"loss": 1.3946,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.12071330589849108,
|
| 776 |
+
"grad_norm": 0.36761683225631714,
|
| 777 |
+
"learning_rate": 1.9753390875462394e-05,
|
| 778 |
+
"loss": 1.3895,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.12181069958847737,
|
| 783 |
+
"grad_norm": 0.3479093313217163,
|
| 784 |
+
"learning_rate": 1.9728729963008633e-05,
|
| 785 |
+
"loss": 1.3704,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.12290809327846365,
|
| 790 |
+
"grad_norm": 0.3511699438095093,
|
| 791 |
+
"learning_rate": 1.9704069050554872e-05,
|
| 792 |
+
"loss": 1.4554,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.12400548696844993,
|
| 797 |
+
"grad_norm": 0.3108881413936615,
|
| 798 |
+
"learning_rate": 1.967940813810111e-05,
|
| 799 |
+
"loss": 1.4446,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.12510288065843622,
|
| 804 |
+
"grad_norm": 0.35024315118789673,
|
| 805 |
+
"learning_rate": 1.965474722564735e-05,
|
| 806 |
+
"loss": 1.3689,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.1262002743484225,
|
| 811 |
+
"grad_norm": 0.3867047429084778,
|
| 812 |
+
"learning_rate": 1.9630086313193592e-05,
|
| 813 |
+
"loss": 1.323,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.12729766803840878,
|
| 818 |
+
"grad_norm": 0.3047441244125366,
|
| 819 |
+
"learning_rate": 1.9605425400739828e-05,
|
| 820 |
+
"loss": 1.3858,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.12839506172839507,
|
| 825 |
+
"grad_norm": 0.34278514981269836,
|
| 826 |
+
"learning_rate": 1.9580764488286066e-05,
|
| 827 |
+
"loss": 1.4294,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.12949245541838134,
|
| 832 |
+
"grad_norm": 0.3425813317298889,
|
| 833 |
+
"learning_rate": 1.955610357583231e-05,
|
| 834 |
+
"loss": 1.3426,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.13058984910836763,
|
| 839 |
+
"grad_norm": 0.38938474655151367,
|
| 840 |
+
"learning_rate": 1.9531442663378544e-05,
|
| 841 |
+
"loss": 1.3857,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.13168724279835392,
|
| 846 |
+
"grad_norm": 0.323210746049881,
|
| 847 |
+
"learning_rate": 1.9506781750924787e-05,
|
| 848 |
+
"loss": 1.3922,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.13278463648834019,
|
| 853 |
+
"grad_norm": 0.33866652846336365,
|
| 854 |
+
"learning_rate": 1.9482120838471025e-05,
|
| 855 |
+
"loss": 1.3969,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.13388203017832648,
|
| 860 |
+
"grad_norm": 0.3292064368724823,
|
| 861 |
+
"learning_rate": 1.9457459926017264e-05,
|
| 862 |
+
"loss": 1.417,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.13497942386831277,
|
| 867 |
+
"grad_norm": 0.3402523100376129,
|
| 868 |
+
"learning_rate": 1.9432799013563503e-05,
|
| 869 |
+
"loss": 1.3128,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.13607681755829903,
|
| 874 |
+
"grad_norm": 0.3458561599254608,
|
| 875 |
+
"learning_rate": 1.9408138101109742e-05,
|
| 876 |
+
"loss": 1.3259,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.13717421124828533,
|
| 881 |
+
"grad_norm": 0.3515847325325012,
|
| 882 |
+
"learning_rate": 1.938347718865598e-05,
|
| 883 |
+
"loss": 1.3399,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.1382716049382716,
|
| 888 |
+
"grad_norm": 0.3250204622745514,
|
| 889 |
+
"learning_rate": 1.935881627620222e-05,
|
| 890 |
+
"loss": 1.325,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.13936899862825788,
|
| 895 |
+
"grad_norm": 0.37784844636917114,
|
| 896 |
+
"learning_rate": 1.9334155363748462e-05,
|
| 897 |
+
"loss": 1.3475,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.14046639231824418,
|
| 902 |
+
"grad_norm": 0.39530113339424133,
|
| 903 |
+
"learning_rate": 1.9309494451294698e-05,
|
| 904 |
+
"loss": 1.3755,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.14156378600823044,
|
| 909 |
+
"grad_norm": 0.37158292531967163,
|
| 910 |
+
"learning_rate": 1.928483353884094e-05,
|
| 911 |
+
"loss": 1.3443,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.14266117969821673,
|
| 916 |
+
"grad_norm": 0.43860819935798645,
|
| 917 |
+
"learning_rate": 1.9260172626387176e-05,
|
| 918 |
+
"loss": 1.3849,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.14375857338820303,
|
| 923 |
+
"grad_norm": 0.43812137842178345,
|
| 924 |
+
"learning_rate": 1.9235511713933418e-05,
|
| 925 |
+
"loss": 1.335,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.1448559670781893,
|
| 930 |
+
"grad_norm": 0.4332147538661957,
|
| 931 |
+
"learning_rate": 1.9210850801479657e-05,
|
| 932 |
+
"loss": 1.286,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.14595336076817558,
|
| 937 |
+
"grad_norm": 0.3903179168701172,
|
| 938 |
+
"learning_rate": 1.9186189889025896e-05,
|
| 939 |
+
"loss": 1.3319,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.14705075445816188,
|
| 944 |
+
"grad_norm": 0.4002543091773987,
|
| 945 |
+
"learning_rate": 1.9161528976572135e-05,
|
| 946 |
+
"loss": 1.384,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.14814814814814814,
|
| 951 |
+
"grad_norm": 0.3899374008178711,
|
| 952 |
+
"learning_rate": 1.9136868064118374e-05,
|
| 953 |
+
"loss": 1.4108,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.14924554183813443,
|
| 958 |
+
"grad_norm": 0.4650733172893524,
|
| 959 |
+
"learning_rate": 1.9112207151664612e-05,
|
| 960 |
+
"loss": 1.3531,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.15034293552812072,
|
| 965 |
+
"grad_norm": 0.4140302538871765,
|
| 966 |
+
"learning_rate": 1.908754623921085e-05,
|
| 967 |
+
"loss": 1.3683,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.151440329218107,
|
| 972 |
+
"grad_norm": 0.46601012349128723,
|
| 973 |
+
"learning_rate": 1.9062885326757094e-05,
|
| 974 |
+
"loss": 1.3669,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.15253772290809328,
|
| 979 |
+
"grad_norm": 0.4634631276130676,
|
| 980 |
+
"learning_rate": 1.903822441430333e-05,
|
| 981 |
+
"loss": 1.3653,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.15363511659807957,
|
| 986 |
+
"grad_norm": 0.49095258116722107,
|
| 987 |
+
"learning_rate": 1.901356350184957e-05,
|
| 988 |
+
"loss": 1.2959,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.15473251028806584,
|
| 993 |
+
"grad_norm": 0.504693865776062,
|
| 994 |
+
"learning_rate": 1.898890258939581e-05,
|
| 995 |
+
"loss": 1.3398,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.15582990397805213,
|
| 1000 |
+
"grad_norm": 0.4257521629333496,
|
| 1001 |
+
"learning_rate": 1.896424167694205e-05,
|
| 1002 |
+
"loss": 1.4279,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.1569272976680384,
|
| 1007 |
+
"grad_norm": 0.4088304936885834,
|
| 1008 |
+
"learning_rate": 1.8939580764488288e-05,
|
| 1009 |
+
"loss": 1.349,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.1580246913580247,
|
| 1014 |
+
"grad_norm": 0.46413642168045044,
|
| 1015 |
+
"learning_rate": 1.8914919852034527e-05,
|
| 1016 |
+
"loss": 1.225,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.15912208504801098,
|
| 1021 |
+
"grad_norm": 0.42032694816589355,
|
| 1022 |
+
"learning_rate": 1.8890258939580766e-05,
|
| 1023 |
+
"loss": 1.3018,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.16021947873799725,
|
| 1028 |
+
"grad_norm": 0.3075321614742279,
|
| 1029 |
+
"learning_rate": 1.8865598027127005e-05,
|
| 1030 |
+
"loss": 1.3478,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.16131687242798354,
|
| 1035 |
+
"grad_norm": 0.2494765967130661,
|
| 1036 |
+
"learning_rate": 1.8840937114673244e-05,
|
| 1037 |
+
"loss": 1.2855,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.16241426611796983,
|
| 1042 |
+
"grad_norm": 0.22226963937282562,
|
| 1043 |
+
"learning_rate": 1.8816276202219483e-05,
|
| 1044 |
+
"loss": 1.2765,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.1635116598079561,
|
| 1049 |
+
"grad_norm": 0.19540823996067047,
|
| 1050 |
+
"learning_rate": 1.879161528976572e-05,
|
| 1051 |
+
"loss": 1.3324,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.1646090534979424,
|
| 1056 |
+
"grad_norm": 0.1837640106678009,
|
| 1057 |
+
"learning_rate": 1.876695437731196e-05,
|
| 1058 |
+
"loss": 1.357,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.16570644718792868,
|
| 1063 |
+
"grad_norm": 0.21854494512081146,
|
| 1064 |
+
"learning_rate": 1.87422934648582e-05,
|
| 1065 |
+
"loss": 1.3023,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.16680384087791494,
|
| 1070 |
+
"grad_norm": 0.1981099396944046,
|
| 1071 |
+
"learning_rate": 1.8717632552404442e-05,
|
| 1072 |
+
"loss": 1.2727,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.16790123456790124,
|
| 1077 |
+
"grad_norm": 0.1930709034204483,
|
| 1078 |
+
"learning_rate": 1.8692971639950677e-05,
|
| 1079 |
+
"loss": 1.309,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.16899862825788753,
|
| 1084 |
+
"grad_norm": 0.196756511926651,
|
| 1085 |
+
"learning_rate": 1.866831072749692e-05,
|
| 1086 |
+
"loss": 1.3215,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.1700960219478738,
|
| 1091 |
+
"grad_norm": 0.18768474459648132,
|
| 1092 |
+
"learning_rate": 1.864364981504316e-05,
|
| 1093 |
+
"loss": 1.3032,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.17119341563786009,
|
| 1098 |
+
"grad_norm": 0.22273507714271545,
|
| 1099 |
+
"learning_rate": 1.8618988902589397e-05,
|
| 1100 |
+
"loss": 1.312,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.17229080932784638,
|
| 1105 |
+
"grad_norm": 0.17846502363681793,
|
| 1106 |
+
"learning_rate": 1.8594327990135636e-05,
|
| 1107 |
+
"loss": 1.3079,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.17338820301783264,
|
| 1112 |
+
"grad_norm": 0.18322448432445526,
|
| 1113 |
+
"learning_rate": 1.8569667077681875e-05,
|
| 1114 |
+
"loss": 1.3537,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.17448559670781894,
|
| 1119 |
+
"grad_norm": 0.180659681558609,
|
| 1120 |
+
"learning_rate": 1.8545006165228114e-05,
|
| 1121 |
+
"loss": 1.32,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.1755829903978052,
|
| 1126 |
+
"grad_norm": 0.23424433171749115,
|
| 1127 |
+
"learning_rate": 1.8520345252774353e-05,
|
| 1128 |
+
"loss": 1.3525,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.1766803840877915,
|
| 1133 |
+
"grad_norm": 0.19839544594287872,
|
| 1134 |
+
"learning_rate": 1.8495684340320595e-05,
|
| 1135 |
+
"loss": 1.2567,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.17777777777777778,
|
| 1140 |
+
"grad_norm": 0.1809428632259369,
|
| 1141 |
+
"learning_rate": 1.847102342786683e-05,
|
| 1142 |
+
"loss": 1.4336,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.17887517146776405,
|
| 1147 |
+
"grad_norm": 0.18282510340213776,
|
| 1148 |
+
"learning_rate": 1.8446362515413073e-05,
|
| 1149 |
+
"loss": 1.3281,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.17997256515775034,
|
| 1154 |
+
"grad_norm": 0.18329757452011108,
|
| 1155 |
+
"learning_rate": 1.8421701602959312e-05,
|
| 1156 |
+
"loss": 1.2928,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.18106995884773663,
|
| 1161 |
+
"grad_norm": 0.1925593912601471,
|
| 1162 |
+
"learning_rate": 1.839704069050555e-05,
|
| 1163 |
+
"loss": 1.3299,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.1821673525377229,
|
| 1168 |
+
"grad_norm": 0.19288820028305054,
|
| 1169 |
+
"learning_rate": 1.837237977805179e-05,
|
| 1170 |
+
"loss": 1.3251,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.1832647462277092,
|
| 1175 |
+
"grad_norm": 0.18203546106815338,
|
| 1176 |
+
"learning_rate": 1.834771886559803e-05,
|
| 1177 |
+
"loss": 1.336,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.18436213991769548,
|
| 1182 |
+
"grad_norm": 0.20783405005931854,
|
| 1183 |
+
"learning_rate": 1.8323057953144268e-05,
|
| 1184 |
+
"loss": 1.3015,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.18545953360768175,
|
| 1189 |
+
"grad_norm": 0.16650988161563873,
|
| 1190 |
+
"learning_rate": 1.8298397040690507e-05,
|
| 1191 |
+
"loss": 1.3584,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.18655692729766804,
|
| 1196 |
+
"grad_norm": 0.19607752561569214,
|
| 1197 |
+
"learning_rate": 1.8273736128236746e-05,
|
| 1198 |
+
"loss": 1.3404,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.18765432098765433,
|
| 1203 |
+
"grad_norm": 0.1942966729402542,
|
| 1204 |
+
"learning_rate": 1.8249075215782984e-05,
|
| 1205 |
+
"loss": 1.3222,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.1887517146776406,
|
| 1210 |
+
"grad_norm": 0.23658064007759094,
|
| 1211 |
+
"learning_rate": 1.8224414303329227e-05,
|
| 1212 |
+
"loss": 1.3226,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.1898491083676269,
|
| 1217 |
+
"grad_norm": 0.1818239539861679,
|
| 1218 |
+
"learning_rate": 1.8199753390875462e-05,
|
| 1219 |
+
"loss": 1.378,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.19094650205761318,
|
| 1224 |
+
"grad_norm": 0.1911785900592804,
|
| 1225 |
+
"learning_rate": 1.8175092478421705e-05,
|
| 1226 |
+
"loss": 1.4018,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.19204389574759945,
|
| 1231 |
+
"grad_norm": 0.20029093325138092,
|
| 1232 |
+
"learning_rate": 1.8150431565967943e-05,
|
| 1233 |
+
"loss": 1.3754,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.19314128943758574,
|
| 1238 |
+
"grad_norm": 0.17715874314308167,
|
| 1239 |
+
"learning_rate": 1.8125770653514182e-05,
|
| 1240 |
+
"loss": 1.2913,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.194238683127572,
|
| 1245 |
+
"grad_norm": 0.19813261926174164,
|
| 1246 |
+
"learning_rate": 1.810110974106042e-05,
|
| 1247 |
+
"loss": 1.4297,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.1953360768175583,
|
| 1252 |
+
"grad_norm": 0.18769969046115875,
|
| 1253 |
+
"learning_rate": 1.807644882860666e-05,
|
| 1254 |
+
"loss": 1.3742,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.1964334705075446,
|
| 1259 |
+
"grad_norm": 0.19279271364212036,
|
| 1260 |
+
"learning_rate": 1.80517879161529e-05,
|
| 1261 |
+
"loss": 1.2959,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.19753086419753085,
|
| 1266 |
+
"grad_norm": 0.18196497857570648,
|
| 1267 |
+
"learning_rate": 1.8027127003699138e-05,
|
| 1268 |
+
"loss": 1.257,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.19862825788751715,
|
| 1273 |
+
"grad_norm": 0.18429674208164215,
|
| 1274 |
+
"learning_rate": 1.8002466091245377e-05,
|
| 1275 |
+
"loss": 1.3139,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.19972565157750344,
|
| 1280 |
+
"grad_norm": 0.17658454179763794,
|
| 1281 |
+
"learning_rate": 1.7977805178791616e-05,
|
| 1282 |
+
"loss": 1.2467,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.2008230452674897,
|
| 1287 |
+
"grad_norm": 0.18059629201889038,
|
| 1288 |
+
"learning_rate": 1.7953144266337855e-05,
|
| 1289 |
+
"loss": 1.3461,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.201920438957476,
|
| 1294 |
+
"grad_norm": 0.17834338545799255,
|
| 1295 |
+
"learning_rate": 1.7928483353884094e-05,
|
| 1296 |
+
"loss": 1.3542,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.2030178326474623,
|
| 1301 |
+
"grad_norm": 0.19244344532489777,
|
| 1302 |
+
"learning_rate": 1.7903822441430333e-05,
|
| 1303 |
+
"loss": 1.3086,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.20411522633744855,
|
| 1308 |
+
"grad_norm": 0.19486865401268005,
|
| 1309 |
+
"learning_rate": 1.7879161528976575e-05,
|
| 1310 |
+
"loss": 1.2946,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.20521262002743484,
|
| 1315 |
+
"grad_norm": 0.19326968491077423,
|
| 1316 |
+
"learning_rate": 1.785450061652281e-05,
|
| 1317 |
+
"loss": 1.2794,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.20631001371742114,
|
| 1322 |
+
"grad_norm": 0.2171463817358017,
|
| 1323 |
+
"learning_rate": 1.7829839704069053e-05,
|
| 1324 |
+
"loss": 1.3009,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.2074074074074074,
|
| 1329 |
+
"grad_norm": 0.2024574726819992,
|
| 1330 |
+
"learning_rate": 1.780517879161529e-05,
|
| 1331 |
+
"loss": 1.3694,
|
| 1332 |
+
"step": 189
|
| 1333 |
}
|
| 1334 |
],
|
| 1335 |
"logging_steps": 1,
|
| 1336 |
+
"max_steps": 911,
|
| 1337 |
"num_input_tokens_seen": 0,
|
| 1338 |
"num_train_epochs": 1,
|
| 1339 |
"save_steps": 1,
|
|
|
|
| 1349 |
"attributes": {}
|
| 1350 |
}
|
| 1351 |
},
|
| 1352 |
+
"total_flos": 1.1524363938073805e+17,
|
| 1353 |
"train_batch_size": 4,
|
| 1354 |
"trial_name": null,
|
| 1355 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3925e6f4fb74ea74c296487c97f50eee65db504b312a1431f8d935775889ba02
|
| 3 |
size 5240
|