Training in progress, step 248, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 144805440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58100fe655c8ed92d4518fef45e29df4d60c7e291f66f716456f73e4ea77f392
|
| 3 |
size 144805440
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 74291604
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab152d90f6d47ebd8a356ceecca0993fb077dcff867c7c58ca00456e2cfcd04b
|
| 3 |
size 74291604
|
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:f660891b19594633ced246d59eedd400fe2556d319f4e5ca333df7fb57888180
|
| 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 sha256:81771ff96e80b84ed048126e169640f8617ceb476fe2f91b8561190057e53b0d
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 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,
|
|
@@ -875,6 +875,874 @@
|
|
| 875 |
"learning_rate": 8.61933911810608e-05,
|
| 876 |
"loss": 0.5258,
|
| 877 |
"step": 124
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 878 |
}
|
| 879 |
],
|
| 880 |
"logging_steps": 1,
|
|
@@ -894,7 +1762,7 @@
|
|
| 894 |
"attributes": {}
|
| 895 |
}
|
| 896 |
},
|
| 897 |
-
"total_flos":
|
| 898 |
"train_batch_size": 4,
|
| 899 |
"trial_name": null,
|
| 900 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5002521432173475,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 248,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 875 |
"learning_rate": 8.61933911810608e-05,
|
| 876 |
"loss": 0.5258,
|
| 877 |
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.2521432173474534,
|
| 881 |
+
"grad_norm": 0.4010732173919678,
|
| 882 |
+
"learning_rate": 8.597192817707122e-05,
|
| 883 |
+
"loss": 0.6078,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.25416036308623297,
|
| 888 |
+
"grad_norm": 0.3758286237716675,
|
| 889 |
+
"learning_rate": 8.574899252492833e-05,
|
| 890 |
+
"loss": 0.6296,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.2561775088250126,
|
| 895 |
+
"grad_norm": 0.4148157238960266,
|
| 896 |
+
"learning_rate": 8.552459335135381e-05,
|
| 897 |
+
"loss": 0.5566,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.25819465456379226,
|
| 902 |
+
"grad_norm": 0.45975151658058167,
|
| 903 |
+
"learning_rate": 8.529873984298418e-05,
|
| 904 |
+
"loss": 0.7839,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.26021180030257185,
|
| 909 |
+
"grad_norm": 0.3960277736186981,
|
| 910 |
+
"learning_rate": 8.507144124599467e-05,
|
| 911 |
+
"loss": 0.5523,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.2622289460413515,
|
| 916 |
+
"grad_norm": 0.42408132553100586,
|
| 917 |
+
"learning_rate": 8.484270686572078e-05,
|
| 918 |
+
"loss": 0.6672,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.2642460917801311,
|
| 923 |
+
"grad_norm": 0.42156749963760376,
|
| 924 |
+
"learning_rate": 8.46125460662772e-05,
|
| 925 |
+
"loss": 0.6694,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.26626323751891073,
|
| 930 |
+
"grad_norm": 0.4371163547039032,
|
| 931 |
+
"learning_rate": 8.43809682701746e-05,
|
| 932 |
+
"loss": 0.6765,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.2682803832576904,
|
| 937 |
+
"grad_norm": 0.4190014600753784,
|
| 938 |
+
"learning_rate": 8.41479829579338e-05,
|
| 939 |
+
"loss": 0.5559,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.27029752899646997,
|
| 944 |
+
"grad_norm": 0.44778770208358765,
|
| 945 |
+
"learning_rate": 8.391359966769765e-05,
|
| 946 |
+
"loss": 0.6315,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.2723146747352496,
|
| 951 |
+
"grad_norm": 0.4291159212589264,
|
| 952 |
+
"learning_rate": 8.367782799484057e-05,
|
| 953 |
+
"loss": 0.6501,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.27433182047402926,
|
| 958 |
+
"grad_norm": 0.4089667797088623,
|
| 959 |
+
"learning_rate": 8.344067759157572e-05,
|
| 960 |
+
"loss": 0.6162,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.27634896621280886,
|
| 965 |
+
"grad_norm": 0.4689813256263733,
|
| 966 |
+
"learning_rate": 8.320215816655987e-05,
|
| 967 |
+
"loss": 0.7768,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.2783661119515885,
|
| 972 |
+
"grad_norm": 0.41810306906700134,
|
| 973 |
+
"learning_rate": 8.296227948449589e-05,
|
| 974 |
+
"loss": 0.5548,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.28038325769036815,
|
| 979 |
+
"grad_norm": 0.4520469009876251,
|
| 980 |
+
"learning_rate": 8.272105136573303e-05,
|
| 981 |
+
"loss": 0.6789,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.28240040342914774,
|
| 986 |
+
"grad_norm": 0.4686160087585449,
|
| 987 |
+
"learning_rate": 8.24784836858649e-05,
|
| 988 |
+
"loss": 0.6922,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.2844175491679274,
|
| 993 |
+
"grad_norm": 0.5166224837303162,
|
| 994 |
+
"learning_rate": 8.223458637532515e-05,
|
| 995 |
+
"loss": 0.7206,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.28643469490670703,
|
| 1000 |
+
"grad_norm": 0.486186683177948,
|
| 1001 |
+
"learning_rate": 8.198936941898091e-05,
|
| 1002 |
+
"loss": 0.599,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.2884518406454866,
|
| 1007 |
+
"grad_norm": 0.4760117530822754,
|
| 1008 |
+
"learning_rate": 8.174284285572408e-05,
|
| 1009 |
+
"loss": 0.6445,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.29046898638426627,
|
| 1014 |
+
"grad_norm": 0.49186456203460693,
|
| 1015 |
+
"learning_rate": 8.14950167780603e-05,
|
| 1016 |
+
"loss": 0.635,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.2924861321230459,
|
| 1021 |
+
"grad_norm": 0.5166196227073669,
|
| 1022 |
+
"learning_rate": 8.12459013316958e-05,
|
| 1023 |
+
"loss": 0.6888,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.2945032778618255,
|
| 1028 |
+
"grad_norm": 0.5494632720947266,
|
| 1029 |
+
"learning_rate": 8.099550671512202e-05,
|
| 1030 |
+
"loss": 0.6592,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.29652042360060515,
|
| 1035 |
+
"grad_norm": 0.5840097069740295,
|
| 1036 |
+
"learning_rate": 8.074384317919812e-05,
|
| 1037 |
+
"loss": 0.7137,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.29853756933938475,
|
| 1042 |
+
"grad_norm": 0.5433554649353027,
|
| 1043 |
+
"learning_rate": 8.049092102673135e-05,
|
| 1044 |
+
"loss": 0.6446,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3005547150781644,
|
| 1049 |
+
"grad_norm": 0.5587112903594971,
|
| 1050 |
+
"learning_rate": 8.023675061205519e-05,
|
| 1051 |
+
"loss": 0.6222,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.30257186081694404,
|
| 1056 |
+
"grad_norm": 0.6611265540122986,
|
| 1057 |
+
"learning_rate": 7.998134234060551e-05,
|
| 1058 |
+
"loss": 0.7365,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.30458900655572363,
|
| 1063 |
+
"grad_norm": 0.49480316042900085,
|
| 1064 |
+
"learning_rate": 7.972470666849457e-05,
|
| 1065 |
+
"loss": 0.5276,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.3066061522945033,
|
| 1070 |
+
"grad_norm": 0.5728369951248169,
|
| 1071 |
+
"learning_rate": 7.946685410208296e-05,
|
| 1072 |
+
"loss": 0.6442,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.3086232980332829,
|
| 1077 |
+
"grad_norm": 0.4378463923931122,
|
| 1078 |
+
"learning_rate": 7.920779519754948e-05,
|
| 1079 |
+
"loss": 0.6289,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.3106404437720625,
|
| 1084 |
+
"grad_norm": 0.42006805539131165,
|
| 1085 |
+
"learning_rate": 7.894754056045901e-05,
|
| 1086 |
+
"loss": 0.545,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.31265758951084216,
|
| 1091 |
+
"grad_norm": 0.4681999683380127,
|
| 1092 |
+
"learning_rate": 7.868610084532828e-05,
|
| 1093 |
+
"loss": 0.5703,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.3146747352496218,
|
| 1098 |
+
"grad_norm": 0.537784218788147,
|
| 1099 |
+
"learning_rate": 7.842348675518968e-05,
|
| 1100 |
+
"loss": 0.6163,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.3166918809884014,
|
| 1105 |
+
"grad_norm": 0.4741944372653961,
|
| 1106 |
+
"learning_rate": 7.815970904115319e-05,
|
| 1107 |
+
"loss": 0.6115,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.31870902672718104,
|
| 1112 |
+
"grad_norm": 0.42608484625816345,
|
| 1113 |
+
"learning_rate": 7.789477850196614e-05,
|
| 1114 |
+
"loss": 0.5959,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.3207261724659607,
|
| 1119 |
+
"grad_norm": 0.44388729333877563,
|
| 1120 |
+
"learning_rate": 7.762870598357115e-05,
|
| 1121 |
+
"loss": 0.5518,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.3227433182047403,
|
| 1126 |
+
"grad_norm": 0.3990045189857483,
|
| 1127 |
+
"learning_rate": 7.736150237866213e-05,
|
| 1128 |
+
"loss": 0.5256,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.32476046394351993,
|
| 1133 |
+
"grad_norm": 1.0403790473937988,
|
| 1134 |
+
"learning_rate": 7.709317862623833e-05,
|
| 1135 |
+
"loss": 0.5994,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.3267776096822995,
|
| 1140 |
+
"grad_norm": 0.38422009348869324,
|
| 1141 |
+
"learning_rate": 7.682374571115651e-05,
|
| 1142 |
+
"loss": 0.5976,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.32879475542107917,
|
| 1147 |
+
"grad_norm": 0.3627714216709137,
|
| 1148 |
+
"learning_rate": 7.655321466368126e-05,
|
| 1149 |
+
"loss": 0.6134,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.3308119011598588,
|
| 1154 |
+
"grad_norm": 0.4387800097465515,
|
| 1155 |
+
"learning_rate": 7.628159655903336e-05,
|
| 1156 |
+
"loss": 0.6016,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.3328290468986384,
|
| 1161 |
+
"grad_norm": 0.38703906536102295,
|
| 1162 |
+
"learning_rate": 7.600890251693645e-05,
|
| 1163 |
+
"loss": 0.5804,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.33484619263741805,
|
| 1168 |
+
"grad_norm": 0.37045204639434814,
|
| 1169 |
+
"learning_rate": 7.57351437011618e-05,
|
| 1170 |
+
"loss": 0.5176,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.3368633383761977,
|
| 1175 |
+
"grad_norm": 0.37859445810317993,
|
| 1176 |
+
"learning_rate": 7.546033131907122e-05,
|
| 1177 |
+
"loss": 0.5251,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.3388804841149773,
|
| 1182 |
+
"grad_norm": 0.3897157907485962,
|
| 1183 |
+
"learning_rate": 7.51844766211583e-05,
|
| 1184 |
+
"loss": 0.5355,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.34089762985375693,
|
| 1189 |
+
"grad_norm": 0.384641170501709,
|
| 1190 |
+
"learning_rate": 7.490759090058778e-05,
|
| 1191 |
+
"loss": 0.5872,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.3429147755925366,
|
| 1196 |
+
"grad_norm": 0.3863964378833771,
|
| 1197 |
+
"learning_rate": 7.462968549273326e-05,
|
| 1198 |
+
"loss": 0.5901,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.34493192133131617,
|
| 1203 |
+
"grad_norm": 0.3983001708984375,
|
| 1204 |
+
"learning_rate": 7.435077177471315e-05,
|
| 1205 |
+
"loss": 0.5495,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.3469490670700958,
|
| 1210 |
+
"grad_norm": 0.419036865234375,
|
| 1211 |
+
"learning_rate": 7.407086116492484e-05,
|
| 1212 |
+
"loss": 0.5547,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.34896621280887546,
|
| 1217 |
+
"grad_norm": 0.39494702219963074,
|
| 1218 |
+
"learning_rate": 7.378996512257735e-05,
|
| 1219 |
+
"loss": 0.5727,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.35098335854765506,
|
| 1224 |
+
"grad_norm": 0.3954225182533264,
|
| 1225 |
+
"learning_rate": 7.35080951472221e-05,
|
| 1226 |
+
"loss": 0.5669,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.3530005042864347,
|
| 1231 |
+
"grad_norm": 0.43136945366859436,
|
| 1232 |
+
"learning_rate": 7.322526277828216e-05,
|
| 1233 |
+
"loss": 0.6174,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.35501765002521435,
|
| 1238 |
+
"grad_norm": 0.37220558524131775,
|
| 1239 |
+
"learning_rate": 7.294147959457989e-05,
|
| 1240 |
+
"loss": 0.5389,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.35703479576399394,
|
| 1245 |
+
"grad_norm": 0.40839049220085144,
|
| 1246 |
+
"learning_rate": 7.265675721386285e-05,
|
| 1247 |
+
"loss": 0.6032,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.3590519415027736,
|
| 1252 |
+
"grad_norm": 0.37361785769462585,
|
| 1253 |
+
"learning_rate": 7.237110729232825e-05,
|
| 1254 |
+
"loss": 0.5343,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.3610690872415532,
|
| 1259 |
+
"grad_norm": 0.4027640223503113,
|
| 1260 |
+
"learning_rate": 7.208454152414571e-05,
|
| 1261 |
+
"loss": 0.5687,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.3630862329803328,
|
| 1266 |
+
"grad_norm": 0.4593392014503479,
|
| 1267 |
+
"learning_rate": 7.179707164097851e-05,
|
| 1268 |
+
"loss": 0.633,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.36510337871911247,
|
| 1273 |
+
"grad_norm": 0.3781510889530182,
|
| 1274 |
+
"learning_rate": 7.150870941150336e-05,
|
| 1275 |
+
"loss": 0.5225,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.36712052445789206,
|
| 1280 |
+
"grad_norm": 0.40543264150619507,
|
| 1281 |
+
"learning_rate": 7.12194666409285e-05,
|
| 1282 |
+
"loss": 0.5756,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.3691376701966717,
|
| 1287 |
+
"grad_norm": 0.42058733105659485,
|
| 1288 |
+
"learning_rate": 7.092935517051058e-05,
|
| 1289 |
+
"loss": 0.6274,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.37115481593545135,
|
| 1294 |
+
"grad_norm": 0.4416365623474121,
|
| 1295 |
+
"learning_rate": 7.063838687706971e-05,
|
| 1296 |
+
"loss": 0.6606,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.37317196167423095,
|
| 1301 |
+
"grad_norm": 0.4479544460773468,
|
| 1302 |
+
"learning_rate": 7.034657367250337e-05,
|
| 1303 |
+
"loss": 0.5858,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.3751891074130106,
|
| 1308 |
+
"grad_norm": 0.45392027497291565,
|
| 1309 |
+
"learning_rate": 7.005392750329868e-05,
|
| 1310 |
+
"loss": 0.6059,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.37720625315179024,
|
| 1315 |
+
"grad_norm": 0.4504959285259247,
|
| 1316 |
+
"learning_rate": 6.976046035004335e-05,
|
| 1317 |
+
"loss": 0.5292,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.37922339889056983,
|
| 1322 |
+
"grad_norm": 0.4504992365837097,
|
| 1323 |
+
"learning_rate": 6.946618422693521e-05,
|
| 1324 |
+
"loss": 0.5815,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.3812405446293495,
|
| 1329 |
+
"grad_norm": 0.44799065589904785,
|
| 1330 |
+
"learning_rate": 6.917111118129035e-05,
|
| 1331 |
+
"loss": 0.5636,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.3832576903681291,
|
| 1336 |
+
"grad_norm": 0.49823272228240967,
|
| 1337 |
+
"learning_rate": 6.887525329304994e-05,
|
| 1338 |
+
"loss": 0.7133,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.3852748361069087,
|
| 1343 |
+
"grad_norm": 0.50130295753479,
|
| 1344 |
+
"learning_rate": 6.857862267428563e-05,
|
| 1345 |
+
"loss": 0.6892,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.38729198184568836,
|
| 1350 |
+
"grad_norm": 0.47600454092025757,
|
| 1351 |
+
"learning_rate": 6.828123146870383e-05,
|
| 1352 |
+
"loss": 0.5957,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.38930912758446795,
|
| 1357 |
+
"grad_norm": 0.44867298007011414,
|
| 1358 |
+
"learning_rate": 6.79830918511484e-05,
|
| 1359 |
+
"loss": 0.6613,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.3913262733232476,
|
| 1364 |
+
"grad_norm": 0.4927201569080353,
|
| 1365 |
+
"learning_rate": 6.76842160271023e-05,
|
| 1366 |
+
"loss": 0.6486,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.39334341906202724,
|
| 1371 |
+
"grad_norm": 0.43527671694755554,
|
| 1372 |
+
"learning_rate": 6.738461623218795e-05,
|
| 1373 |
+
"loss": 0.5978,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.39536056480080684,
|
| 1378 |
+
"grad_norm": 0.5360156297683716,
|
| 1379 |
+
"learning_rate": 6.708430473166628e-05,
|
| 1380 |
+
"loss": 0.6932,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.3973777105395865,
|
| 1385 |
+
"grad_norm": 0.5388275384902954,
|
| 1386 |
+
"learning_rate": 6.678329381993458e-05,
|
| 1387 |
+
"loss": 0.6456,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.39939485627836613,
|
| 1392 |
+
"grad_norm": 0.544686496257782,
|
| 1393 |
+
"learning_rate": 6.648159582002322e-05,
|
| 1394 |
+
"loss": 0.6551,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.4014120020171457,
|
| 1399 |
+
"grad_norm": 0.6613960266113281,
|
| 1400 |
+
"learning_rate": 6.617922308309115e-05,
|
| 1401 |
+
"loss": 0.6676,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.40342914775592537,
|
| 1406 |
+
"grad_norm": 0.7599596381187439,
|
| 1407 |
+
"learning_rate": 6.587618798792022e-05,
|
| 1408 |
+
"loss": 0.6209,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.405446293494705,
|
| 1413 |
+
"grad_norm": 0.3868520259857178,
|
| 1414 |
+
"learning_rate": 6.557250294040849e-05,
|
| 1415 |
+
"loss": 0.505,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.4074634392334846,
|
| 1420 |
+
"grad_norm": 0.4449853003025055,
|
| 1421 |
+
"learning_rate": 6.526818037306228e-05,
|
| 1422 |
+
"loss": 0.5904,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.40948058497226425,
|
| 1427 |
+
"grad_norm": 0.45996958017349243,
|
| 1428 |
+
"learning_rate": 6.496323274448721e-05,
|
| 1429 |
+
"loss": 0.543,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.4114977307110439,
|
| 1434 |
+
"grad_norm": 0.49261391162872314,
|
| 1435 |
+
"learning_rate": 6.46576725388782e-05,
|
| 1436 |
+
"loss": 0.6735,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.4135148764498235,
|
| 1441 |
+
"grad_norm": 0.43575477600097656,
|
| 1442 |
+
"learning_rate": 6.435151226550829e-05,
|
| 1443 |
+
"loss": 0.5711,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.41553202218860313,
|
| 1448 |
+
"grad_norm": 0.44983670115470886,
|
| 1449 |
+
"learning_rate": 6.404476445821663e-05,
|
| 1450 |
+
"loss": 0.5746,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.4175491679273828,
|
| 1455 |
+
"grad_norm": 0.3756658136844635,
|
| 1456 |
+
"learning_rate": 6.373744167489531e-05,
|
| 1457 |
+
"loss": 0.5298,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.41956631366616237,
|
| 1462 |
+
"grad_norm": 0.4081321954727173,
|
| 1463 |
+
"learning_rate": 6.342955649697523e-05,
|
| 1464 |
+
"loss": 0.6043,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.421583459404942,
|
| 1469 |
+
"grad_norm": 0.42804038524627686,
|
| 1470 |
+
"learning_rate": 6.312112152891107e-05,
|
| 1471 |
+
"loss": 0.5759,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.4236006051437216,
|
| 1476 |
+
"grad_norm": 0.3999483287334442,
|
| 1477 |
+
"learning_rate": 6.28121493976653e-05,
|
| 1478 |
+
"loss": 0.5183,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.42561775088250126,
|
| 1483 |
+
"grad_norm": 0.40822234749794006,
|
| 1484 |
+
"learning_rate": 6.250265275219116e-05,
|
| 1485 |
+
"loss": 0.6014,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.4276348966212809,
|
| 1490 |
+
"grad_norm": 0.388310045003891,
|
| 1491 |
+
"learning_rate": 6.219264426291494e-05,
|
| 1492 |
+
"loss": 0.5951,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.4296520423600605,
|
| 1497 |
+
"grad_norm": 0.39607641100883484,
|
| 1498 |
+
"learning_rate": 6.188213662121716e-05,
|
| 1499 |
+
"loss": 0.5743,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.43166918809884014,
|
| 1504 |
+
"grad_norm": 0.37904173135757446,
|
| 1505 |
+
"learning_rate": 6.157114253891307e-05,
|
| 1506 |
+
"loss": 0.5331,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.4336863338376198,
|
| 1511 |
+
"grad_norm": 0.36542603373527527,
|
| 1512 |
+
"learning_rate": 6.125967474773223e-05,
|
| 1513 |
+
"loss": 0.5145,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.4357034795763994,
|
| 1518 |
+
"grad_norm": 0.37045159935951233,
|
| 1519 |
+
"learning_rate": 6.0947745998797266e-05,
|
| 1520 |
+
"loss": 0.5091,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.437720625315179,
|
| 1525 |
+
"grad_norm": 0.41620463132858276,
|
| 1526 |
+
"learning_rate": 6.0635369062101875e-05,
|
| 1527 |
+
"loss": 0.6119,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.43973777105395867,
|
| 1532 |
+
"grad_norm": 0.4387664794921875,
|
| 1533 |
+
"learning_rate": 6.032255672598803e-05,
|
| 1534 |
+
"loss": 0.6682,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.44175491679273826,
|
| 1539 |
+
"grad_norm": 0.4237648844718933,
|
| 1540 |
+
"learning_rate": 6.0009321796622444e-05,
|
| 1541 |
+
"loss": 0.5406,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.4437720625315179,
|
| 1546 |
+
"grad_norm": 0.43322858214378357,
|
| 1547 |
+
"learning_rate": 5.969567709747228e-05,
|
| 1548 |
+
"loss": 0.5316,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.44578920827029755,
|
| 1553 |
+
"grad_norm": 0.4600161910057068,
|
| 1554 |
+
"learning_rate": 5.938163546878024e-05,
|
| 1555 |
+
"loss": 0.567,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.44780635400907715,
|
| 1560 |
+
"grad_norm": 0.3887118399143219,
|
| 1561 |
+
"learning_rate": 5.906720976703877e-05,
|
| 1562 |
+
"loss": 0.5762,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.4498234997478568,
|
| 1567 |
+
"grad_norm": 0.4044482409954071,
|
| 1568 |
+
"learning_rate": 5.87524128644639e-05,
|
| 1569 |
+
"loss": 0.6196,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.4518406454866364,
|
| 1574 |
+
"grad_norm": 0.4436338543891907,
|
| 1575 |
+
"learning_rate": 5.843725764846812e-05,
|
| 1576 |
+
"loss": 0.5333,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.45385779122541603,
|
| 1581 |
+
"grad_norm": 0.454662948846817,
|
| 1582 |
+
"learning_rate": 5.812175702113286e-05,
|
| 1583 |
+
"loss": 0.557,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.4558749369641957,
|
| 1588 |
+
"grad_norm": 0.445640504360199,
|
| 1589 |
+
"learning_rate": 5.7805923898680305e-05,
|
| 1590 |
+
"loss": 0.5332,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.45789208270297527,
|
| 1595 |
+
"grad_norm": 0.4665926992893219,
|
| 1596 |
+
"learning_rate": 5.7489771210944564e-05,
|
| 1597 |
+
"loss": 0.6367,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.4599092284417549,
|
| 1602 |
+
"grad_norm": 0.45142456889152527,
|
| 1603 |
+
"learning_rate": 5.717331190084243e-05,
|
| 1604 |
+
"loss": 0.5792,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.46192637418053456,
|
| 1609 |
+
"grad_norm": 0.4474796652793884,
|
| 1610 |
+
"learning_rate": 5.6856558923843364e-05,
|
| 1611 |
+
"loss": 0.575,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.46394351991931415,
|
| 1616 |
+
"grad_norm": 0.4471490681171417,
|
| 1617 |
+
"learning_rate": 5.6539525247439274e-05,
|
| 1618 |
+
"loss": 0.5461,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.4659606656580938,
|
| 1623 |
+
"grad_norm": 0.4579067528247833,
|
| 1624 |
+
"learning_rate": 5.622222385061353e-05,
|
| 1625 |
+
"loss": 0.6183,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.46797781139687344,
|
| 1630 |
+
"grad_norm": 0.4598318636417389,
|
| 1631 |
+
"learning_rate": 5.590466772330968e-05,
|
| 1632 |
+
"loss": 0.6318,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.46999495713565304,
|
| 1637 |
+
"grad_norm": 0.4309539496898651,
|
| 1638 |
+
"learning_rate": 5.558686986589963e-05,
|
| 1639 |
+
"loss": 0.6644,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.4720121028744327,
|
| 1644 |
+
"grad_norm": 0.46067124605178833,
|
| 1645 |
+
"learning_rate": 5.526884328865142e-05,
|
| 1646 |
+
"loss": 0.6401,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.47402924861321233,
|
| 1651 |
+
"grad_norm": 0.450652539730072,
|
| 1652 |
+
"learning_rate": 5.495060101119662e-05,
|
| 1653 |
+
"loss": 0.5475,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.4760463943519919,
|
| 1658 |
+
"grad_norm": 0.44922497868537903,
|
| 1659 |
+
"learning_rate": 5.463215606199733e-05,
|
| 1660 |
+
"loss": 0.6372,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.47806354009077157,
|
| 1665 |
+
"grad_norm": 0.4988616704940796,
|
| 1666 |
+
"learning_rate": 5.431352147781275e-05,
|
| 1667 |
+
"loss": 0.6847,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.4800806858295512,
|
| 1672 |
+
"grad_norm": 0.45364323258399963,
|
| 1673 |
+
"learning_rate": 5.399471030316554e-05,
|
| 1674 |
+
"loss": 0.6203,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.4820978315683308,
|
| 1679 |
+
"grad_norm": 0.4748307466506958,
|
| 1680 |
+
"learning_rate": 5.367573558980775e-05,
|
| 1681 |
+
"loss": 0.571,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.48411497730711045,
|
| 1686 |
+
"grad_norm": 0.46398502588272095,
|
| 1687 |
+
"learning_rate": 5.335661039618653e-05,
|
| 1688 |
+
"loss": 0.6339,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.48613212304589004,
|
| 1693 |
+
"grad_norm": 0.49548378586769104,
|
| 1694 |
+
"learning_rate": 5.3037347786909495e-05,
|
| 1695 |
+
"loss": 0.6616,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.4881492687846697,
|
| 1700 |
+
"grad_norm": 0.4696505665779114,
|
| 1701 |
+
"learning_rate": 5.2717960832209914e-05,
|
| 1702 |
+
"loss": 0.6289,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.49016641452344933,
|
| 1707 |
+
"grad_norm": 0.4907710552215576,
|
| 1708 |
+
"learning_rate": 5.239846260741158e-05,
|
| 1709 |
+
"loss": 0.5494,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.4921835602622289,
|
| 1714 |
+
"grad_norm": 0.522820234298706,
|
| 1715 |
+
"learning_rate": 5.2078866192393574e-05,
|
| 1716 |
+
"loss": 0.6741,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.49420070600100857,
|
| 1721 |
+
"grad_norm": 0.5473442673683167,
|
| 1722 |
+
"learning_rate": 5.1759184671054785e-05,
|
| 1723 |
+
"loss": 0.6574,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.4962178517397882,
|
| 1728 |
+
"grad_norm": 0.5453885197639465,
|
| 1729 |
+
"learning_rate": 5.1439431130778206e-05,
|
| 1730 |
+
"loss": 0.6463,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.4982349974785678,
|
| 1735 |
+
"grad_norm": 0.5924301743507385,
|
| 1736 |
+
"learning_rate": 5.111961866189524e-05,
|
| 1737 |
+
"loss": 0.675,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.5002521432173475,
|
| 1742 |
+
"grad_norm": 0.6364018321037292,
|
| 1743 |
+
"learning_rate": 5.079976035714976e-05,
|
| 1744 |
+
"loss": 0.7198,
|
| 1745 |
+
"step": 248
|
| 1746 |
}
|
| 1747 |
],
|
| 1748 |
"logging_steps": 1,
|
|
|
|
| 1762 |
"attributes": {}
|
| 1763 |
}
|
| 1764 |
},
|
| 1765 |
+
"total_flos": 4.194181341983539e+17,
|
| 1766 |
"train_batch_size": 4,
|
| 1767 |
"trial_name": null,
|
| 1768 |
"trial_params": null
|