Training in progress, epoch 0, checkpoint
Browse files- last-checkpoint/adapter_model.safetensors +1 -1
- last-checkpoint/global_step300/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step300/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step300/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step300/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step300/mp_rank_00_model_states.pt +3 -0
- last-checkpoint/latest +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +512 -4
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 34895376
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:128d5ff87a9a2f67bfc376762663cddc1baec41d7fc66c9f706a34faa11d5ce6
|
| 3 |
size 34895376
|
last-checkpoint/global_step300/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e67dd2c6eafa9e6373d4c602c404f3037feea34cf55ac5d5e5a7db1b2ba46303
|
| 3 |
+
size 26298085
|
last-checkpoint/global_step300/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28595be6eafc5d9e98b7173aa15802b7097913aedc4af74767a7c63f75db51db
|
| 3 |
+
size 26298085
|
last-checkpoint/global_step300/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d9531c02cab629a21f79c760ef90370c1f164696c1a6dc03ed212646d48458a
|
| 3 |
+
size 26298149
|
last-checkpoint/global_step300/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7578bc6c6c3ef05125df821f9f3409b6dbfd2844aba382f7af14c2d20459b4bc
|
| 3 |
+
size 26298149
|
last-checkpoint/global_step300/mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df81a1524d9d57d65c7e452ca66412a0d9fe6e7785143cffdaf5e948a590fc02
|
| 3 |
+
size 501799973
|
last-checkpoint/latest
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
|
|
|
|
| 1 |
+
global_step300
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15429
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dcd295e94e121d0e334ec06742c998279f2c41bf799fbd89390e9e5d7441448
|
| 3 |
size 15429
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:39ae4eb8db6190f758a03220812e9fcbc2a4823df90e3ccf1ba4764886733e3d
|
| 3 |
size 15365
|
last-checkpoint/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15429
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9bbe37b1011b486d46f84f6c3acc88f2b0b90b3a3c2a6f2b01998012c708e63b
|
| 3 |
size 15429
|
last-checkpoint/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15429
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08c2ef4d00af7939a9a25681a8548258d41f42c1471bf03f68a181fcb291b727
|
| 3 |
size 15429
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1401
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:738f1bd52a0e5e3570eb826d97ec615ac0a30012cc827a5a3725e6329285a9f5
|
| 3 |
size 1401
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": null,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -1024,11 +1024,519 @@
|
|
| 1024 |
"eval_samples_per_second": 4.743,
|
| 1025 |
"eval_steps_per_second": 0.308,
|
| 1026 |
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1027 |
}
|
| 1028 |
],
|
| 1029 |
"logging_steps": 5,
|
| 1030 |
"max_steps": 1000,
|
| 1031 |
-
"num_input_tokens_seen":
|
| 1032 |
"num_train_epochs": 2,
|
| 1033 |
"save_steps": 100,
|
| 1034 |
"stateful_callbacks": {
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": null,
|
| 3 |
+
"best_metric": 0.0143232811242342,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.38809831824062097,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 300,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 1024 |
"eval_samples_per_second": 4.743,
|
| 1025 |
"eval_steps_per_second": 0.308,
|
| 1026 |
"step": 200
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"clip_ratio/high_max": 0.0,
|
| 1030 |
+
"clip_ratio/high_mean": 0.0,
|
| 1031 |
+
"clip_ratio/low_mean": 0.0,
|
| 1032 |
+
"clip_ratio/low_min": 0.0,
|
| 1033 |
+
"clip_ratio/region_mean": 0.0,
|
| 1034 |
+
"completions/clipped_ratio": 0.984375,
|
| 1035 |
+
"completions/max_length": 32.0,
|
| 1036 |
+
"completions/max_terminated_length": 11.7,
|
| 1037 |
+
"completions/mean_length": 31.9,
|
| 1038 |
+
"completions/mean_terminated_length": 10.7,
|
| 1039 |
+
"completions/min_length": 28.9,
|
| 1040 |
+
"completions/min_terminated_length": 9.7,
|
| 1041 |
+
"epoch": 0.2652005174644243,
|
| 1042 |
+
"grad_norm": 0.3019338548183441,
|
| 1043 |
+
"kl": 0.37265625,
|
| 1044 |
+
"learning_rate": 0.00019348256763960145,
|
| 1045 |
+
"loss": 0.0152,
|
| 1046 |
+
"num_tokens": 227432.0,
|
| 1047 |
+
"reward": 2.7201414823532106,
|
| 1048 |
+
"reward_std": 0.28786033391952515,
|
| 1049 |
+
"rewards/keyword_inclusion_reward/mean": 0.3062500089406967,
|
| 1050 |
+
"rewards/keyword_inclusion_reward/std": 0.08914534375071526,
|
| 1051 |
+
"step": 205
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"clip_ratio/high_max": 0.0,
|
| 1055 |
+
"clip_ratio/high_mean": 0.0,
|
| 1056 |
+
"clip_ratio/low_mean": 0.0,
|
| 1057 |
+
"clip_ratio/low_min": 0.0,
|
| 1058 |
+
"clip_ratio/region_mean": 0.0,
|
| 1059 |
+
"completions/clipped_ratio": 0.9625,
|
| 1060 |
+
"completions/max_length": 32.0,
|
| 1061 |
+
"completions/max_terminated_length": 17.6,
|
| 1062 |
+
"completions/mean_length": 31.86875,
|
| 1063 |
+
"completions/mean_terminated_length": 16.76666679382324,
|
| 1064 |
+
"completions/min_length": 28.2,
|
| 1065 |
+
"completions/min_terminated_length": 15.4,
|
| 1066 |
+
"epoch": 0.2716688227684347,
|
| 1067 |
+
"grad_norm": 0.09334196895360947,
|
| 1068 |
+
"kl": 0.4015625,
|
| 1069 |
+
"learning_rate": 0.00019284858268809137,
|
| 1070 |
+
"loss": 0.0161,
|
| 1071 |
+
"num_tokens": 232989.0,
|
| 1072 |
+
"reward": 2.8681763648986816,
|
| 1073 |
+
"reward_std": 0.13084560632705688,
|
| 1074 |
+
"rewards/keyword_inclusion_reward/mean": 0.3229166746139526,
|
| 1075 |
+
"rewards/keyword_inclusion_reward/std": 0.04457640573382378,
|
| 1076 |
+
"step": 210
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"clip_ratio/high_max": 0.0,
|
| 1080 |
+
"clip_ratio/high_mean": 0.0,
|
| 1081 |
+
"clip_ratio/low_mean": 0.0,
|
| 1082 |
+
"clip_ratio/low_min": 0.0,
|
| 1083 |
+
"clip_ratio/region_mean": 0.0,
|
| 1084 |
+
"completions/clipped_ratio": 0.91875,
|
| 1085 |
+
"completions/max_length": 32.0,
|
| 1086 |
+
"completions/max_terminated_length": 26.8,
|
| 1087 |
+
"completions/mean_length": 31.3375,
|
| 1088 |
+
"completions/mean_terminated_length": 24.0,
|
| 1089 |
+
"completions/min_length": 21.2,
|
| 1090 |
+
"completions/min_terminated_length": 21.2,
|
| 1091 |
+
"epoch": 0.278137128072445,
|
| 1092 |
+
"grad_norm": 0.1729985624551773,
|
| 1093 |
+
"kl": 0.4,
|
| 1094 |
+
"learning_rate": 0.00019218631515885006,
|
| 1095 |
+
"loss": 0.0126,
|
| 1096 |
+
"num_tokens": 238507.0,
|
| 1097 |
+
"reward": 2.9051851272583007,
|
| 1098 |
+
"reward_std": 0.07850736379623413,
|
| 1099 |
+
"rewards/keyword_inclusion_reward/mean": 0.3270833492279053,
|
| 1100 |
+
"rewards/keyword_inclusion_reward/std": 0.03535533919930458,
|
| 1101 |
+
"step": 215
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"clip_ratio/high_max": 0.0,
|
| 1105 |
+
"clip_ratio/high_mean": 0.0,
|
| 1106 |
+
"clip_ratio/low_mean": 0.0,
|
| 1107 |
+
"clip_ratio/low_min": 0.0,
|
| 1108 |
+
"clip_ratio/region_mean": 0.0,
|
| 1109 |
+
"completions/clipped_ratio": 0.9125,
|
| 1110 |
+
"completions/max_length": 32.0,
|
| 1111 |
+
"completions/max_terminated_length": 24.8,
|
| 1112 |
+
"completions/mean_length": 31.3625,
|
| 1113 |
+
"completions/mean_terminated_length": 20.679999923706056,
|
| 1114 |
+
"completions/min_length": 22.4,
|
| 1115 |
+
"completions/min_terminated_length": 16.0,
|
| 1116 |
+
"epoch": 0.2846054333764554,
|
| 1117 |
+
"grad_norm": 0.21469204127788544,
|
| 1118 |
+
"kl": 0.3677734375,
|
| 1119 |
+
"learning_rate": 0.0001914959667849825,
|
| 1120 |
+
"loss": 0.013,
|
| 1121 |
+
"num_tokens": 244017.0,
|
| 1122 |
+
"reward": 2.8126633167266846,
|
| 1123 |
+
"reward_std": 0.20935297012329102,
|
| 1124 |
+
"rewards/keyword_inclusion_reward/mean": 0.3166666805744171,
|
| 1125 |
+
"rewards/keyword_inclusion_reward/std": 0.0714939571917057,
|
| 1126 |
+
"step": 220
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"clip_ratio/high_max": 0.0,
|
| 1130 |
+
"clip_ratio/high_mean": 0.0,
|
| 1131 |
+
"clip_ratio/low_mean": 0.0,
|
| 1132 |
+
"clip_ratio/low_min": 0.0,
|
| 1133 |
+
"clip_ratio/region_mean": 0.0,
|
| 1134 |
+
"completions/clipped_ratio": 0.93125,
|
| 1135 |
+
"completions/max_length": 32.0,
|
| 1136 |
+
"completions/max_terminated_length": 22.2,
|
| 1137 |
+
"completions/mean_length": 31.6125,
|
| 1138 |
+
"completions/mean_terminated_length": 20.416666793823243,
|
| 1139 |
+
"completions/min_length": 25.4,
|
| 1140 |
+
"completions/min_terminated_length": 19.0,
|
| 1141 |
+
"epoch": 0.2910737386804657,
|
| 1142 |
+
"grad_norm": 0.048688553273677826,
|
| 1143 |
+
"kl": 0.344921875,
|
| 1144 |
+
"learning_rate": 0.00019077774785329087,
|
| 1145 |
+
"loss": 0.0161,
|
| 1146 |
+
"num_tokens": 249609.0,
|
| 1147 |
+
"reward": 2.886680793762207,
|
| 1148 |
+
"reward_std": 0.10467648506164551,
|
| 1149 |
+
"rewards/keyword_inclusion_reward/mean": 0.32500001788139343,
|
| 1150 |
+
"rewards/keyword_inclusion_reward/std": 0.047140452265739444,
|
| 1151 |
+
"step": 225
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"clip_ratio/high_max": 0.0,
|
| 1155 |
+
"clip_ratio/high_mean": 0.0,
|
| 1156 |
+
"clip_ratio/low_mean": 0.0,
|
| 1157 |
+
"clip_ratio/low_min": 0.0,
|
| 1158 |
+
"clip_ratio/region_mean": 0.0,
|
| 1159 |
+
"completions/clipped_ratio": 0.89375,
|
| 1160 |
+
"completions/max_length": 32.0,
|
| 1161 |
+
"completions/max_terminated_length": 28.6,
|
| 1162 |
+
"completions/mean_length": 31.3375,
|
| 1163 |
+
"completions/mean_terminated_length": 25.214286041259765,
|
| 1164 |
+
"completions/min_length": 21.6,
|
| 1165 |
+
"completions/min_terminated_length": 21.6,
|
| 1166 |
+
"epoch": 0.2975420439844761,
|
| 1167 |
+
"grad_norm": 0.03536103293299675,
|
| 1168 |
+
"kl": 0.3490234375,
|
| 1169 |
+
"learning_rate": 0.00019003187714021938,
|
| 1170 |
+
"loss": 0.0157,
|
| 1171 |
+
"num_tokens": 255097.0,
|
| 1172 |
+
"reward": 2.8866806983947755,
|
| 1173 |
+
"reward_std": 0.10467648506164551,
|
| 1174 |
+
"rewards/keyword_inclusion_reward/mean": 0.325000011920929,
|
| 1175 |
+
"rewards/keyword_inclusion_reward/std": 0.03152808472514153,
|
| 1176 |
+
"step": 230
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"clip_ratio/high_max": 0.0,
|
| 1180 |
+
"clip_ratio/high_mean": 0.0,
|
| 1181 |
+
"clip_ratio/low_mean": 0.0,
|
| 1182 |
+
"clip_ratio/low_min": 0.0,
|
| 1183 |
+
"clip_ratio/region_mean": 0.0,
|
| 1184 |
+
"completions/clipped_ratio": 0.8875,
|
| 1185 |
+
"completions/max_length": 32.0,
|
| 1186 |
+
"completions/max_terminated_length": 30.2,
|
| 1187 |
+
"completions/mean_length": 31.24375,
|
| 1188 |
+
"completions/mean_terminated_length": 25.026667022705077,
|
| 1189 |
+
"completions/min_length": 18.0,
|
| 1190 |
+
"completions/min_terminated_length": 18.0,
|
| 1191 |
+
"epoch": 0.3040103492884864,
|
| 1192 |
+
"grad_norm": 0.14790986478328705,
|
| 1193 |
+
"kl": 0.3205078125,
|
| 1194 |
+
"learning_rate": 0.00018925858184521256,
|
| 1195 |
+
"loss": 0.013,
|
| 1196 |
+
"num_tokens": 260542.0,
|
| 1197 |
+
"reward": 2.8126633167266846,
|
| 1198 |
+
"reward_std": 0.20935297012329102,
|
| 1199 |
+
"rewards/keyword_inclusion_reward/mean": 0.3166666805744171,
|
| 1200 |
+
"rewards/keyword_inclusion_reward/std": 0.0714939571917057,
|
| 1201 |
+
"step": 235
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"clip_ratio/high_max": 0.0,
|
| 1205 |
+
"clip_ratio/high_mean": 0.0,
|
| 1206 |
+
"clip_ratio/low_mean": 0.0,
|
| 1207 |
+
"clip_ratio/low_min": 0.0,
|
| 1208 |
+
"clip_ratio/region_mean": 0.0,
|
| 1209 |
+
"completions/clipped_ratio": 0.8875,
|
| 1210 |
+
"completions/max_length": 32.0,
|
| 1211 |
+
"completions/max_terminated_length": 30.2,
|
| 1212 |
+
"completions/mean_length": 31.30625,
|
| 1213 |
+
"completions/mean_terminated_length": 25.746667098999023,
|
| 1214 |
+
"completions/min_length": 20.0,
|
| 1215 |
+
"completions/min_terminated_length": 20.0,
|
| 1216 |
+
"epoch": 0.31047865459249674,
|
| 1217 |
+
"grad_norm": 0.2648606300354004,
|
| 1218 |
+
"kl": 0.3302734375,
|
| 1219 |
+
"learning_rate": 0.0001884580975215084,
|
| 1220 |
+
"loss": 0.0124,
|
| 1221 |
+
"num_tokens": 266065.0,
|
| 1222 |
+
"reward": 2.812663269042969,
|
| 1223 |
+
"reward_std": 0.20935297012329102,
|
| 1224 |
+
"rewards/keyword_inclusion_reward/mean": 0.31666667461395265,
|
| 1225 |
+
"rewards/keyword_inclusion_reward/std": 0.06431937739253044,
|
| 1226 |
+
"step": 240
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"clip_ratio/high_max": 0.0,
|
| 1230 |
+
"clip_ratio/high_mean": 0.0,
|
| 1231 |
+
"clip_ratio/low_mean": 0.0,
|
| 1232 |
+
"clip_ratio/low_min": 0.0,
|
| 1233 |
+
"clip_ratio/region_mean": 0.0,
|
| 1234 |
+
"completions/clipped_ratio": 0.9,
|
| 1235 |
+
"completions/max_length": 32.0,
|
| 1236 |
+
"completions/max_terminated_length": 25.0,
|
| 1237 |
+
"completions/mean_length": 31.53125,
|
| 1238 |
+
"completions/mean_terminated_length": 22.06666717529297,
|
| 1239 |
+
"completions/min_length": 25.2,
|
| 1240 |
+
"completions/min_terminated_length": 18.8,
|
| 1241 |
+
"epoch": 0.3169469598965071,
|
| 1242 |
+
"grad_norm": 0.14858758449554443,
|
| 1243 |
+
"kl": 0.3453125,
|
| 1244 |
+
"learning_rate": 0.00018763066800438636,
|
| 1245 |
+
"loss": 0.0138,
|
| 1246 |
+
"num_tokens": 271536.0,
|
| 1247 |
+
"reward": 2.905185079574585,
|
| 1248 |
+
"reward_std": 0.07850736379623413,
|
| 1249 |
+
"rewards/keyword_inclusion_reward/mean": 0.3270833432674408,
|
| 1250 |
+
"rewards/keyword_inclusion_reward/std": 0.028180759400129318,
|
| 1251 |
+
"step": 245
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"clip_ratio/high_max": 0.0,
|
| 1255 |
+
"clip_ratio/high_mean": 0.0,
|
| 1256 |
+
"clip_ratio/low_mean": 0.0,
|
| 1257 |
+
"clip_ratio/low_min": 0.0,
|
| 1258 |
+
"clip_ratio/region_mean": 0.0,
|
| 1259 |
+
"completions/clipped_ratio": 0.93125,
|
| 1260 |
+
"completions/max_length": 32.0,
|
| 1261 |
+
"completions/max_terminated_length": 30.8,
|
| 1262 |
+
"completions/mean_length": 31.6875,
|
| 1263 |
+
"completions/mean_terminated_length": 28.25,
|
| 1264 |
+
"completions/min_length": 25.6,
|
| 1265 |
+
"completions/min_terminated_length": 25.6,
|
| 1266 |
+
"epoch": 0.32341526520051744,
|
| 1267 |
+
"grad_norm": 0.039299800992012024,
|
| 1268 |
+
"kl": 0.3734375,
|
| 1269 |
+
"learning_rate": 0.00018677654533689287,
|
| 1270 |
+
"loss": 0.0149,
|
| 1271 |
+
"num_tokens": 277104.0,
|
| 1272 |
+
"reward": 2.9236894607543946,
|
| 1273 |
+
"reward_std": 0.052338242530822754,
|
| 1274 |
+
"rewards/keyword_inclusion_reward/mean": 0.3291666805744171,
|
| 1275 |
+
"rewards/keyword_inclusion_reward/std": 0.023570226132869722,
|
| 1276 |
+
"step": 250
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"clip_ratio/high_max": 0.0,
|
| 1280 |
+
"clip_ratio/high_mean": 0.0,
|
| 1281 |
+
"clip_ratio/low_mean": 0.0,
|
| 1282 |
+
"clip_ratio/low_min": 0.0,
|
| 1283 |
+
"clip_ratio/region_mean": 0.0,
|
| 1284 |
+
"completions/clipped_ratio": 0.95,
|
| 1285 |
+
"completions/max_length": 32.0,
|
| 1286 |
+
"completions/max_terminated_length": 22.4,
|
| 1287 |
+
"completions/mean_length": 31.65625,
|
| 1288 |
+
"completions/mean_terminated_length": 20.1,
|
| 1289 |
+
"completions/min_length": 24.2,
|
| 1290 |
+
"completions/min_terminated_length": 17.8,
|
| 1291 |
+
"epoch": 0.3298835705045278,
|
| 1292 |
+
"grad_norm": 0.2771835923194885,
|
| 1293 |
+
"kl": 0.35390625,
|
| 1294 |
+
"learning_rate": 0.00018589598969306645,
|
| 1295 |
+
"loss": 0.0112,
|
| 1296 |
+
"num_tokens": 282653.0,
|
| 1297 |
+
"reward": 2.7941588878631594,
|
| 1298 |
+
"reward_std": 0.28786033391952515,
|
| 1299 |
+
"rewards/keyword_inclusion_reward/mean": 0.3145833432674408,
|
| 1300 |
+
"rewards/keyword_inclusion_reward/std": 0.07305490300059318,
|
| 1301 |
+
"step": 255
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"clip_ratio/high_max": 0.0,
|
| 1305 |
+
"clip_ratio/high_mean": 0.0,
|
| 1306 |
+
"clip_ratio/low_mean": 0.0,
|
| 1307 |
+
"clip_ratio/low_min": 0.0,
|
| 1308 |
+
"clip_ratio/region_mean": 0.0,
|
| 1309 |
+
"completions/clipped_ratio": 0.98125,
|
| 1310 |
+
"completions/max_length": 32.0,
|
| 1311 |
+
"completions/max_terminated_length": 16.8,
|
| 1312 |
+
"completions/mean_length": 31.925,
|
| 1313 |
+
"completions/mean_terminated_length": 16.8,
|
| 1314 |
+
"completions/min_length": 29.6,
|
| 1315 |
+
"completions/min_terminated_length": 16.8,
|
| 1316 |
+
"epoch": 0.33635187580853815,
|
| 1317 |
+
"grad_norm": 0.04248747602105141,
|
| 1318 |
+
"kl": 0.3732421875,
|
| 1319 |
+
"learning_rate": 0.00018498926929868642,
|
| 1320 |
+
"loss": 0.0149,
|
| 1321 |
+
"num_tokens": 288199.0,
|
| 1322 |
+
"reward": 2.8311676025390624,
|
| 1323 |
+
"reward_std": 0.18318384885787964,
|
| 1324 |
+
"rewards/keyword_inclusion_reward/mean": 0.31875001192092894,
|
| 1325 |
+
"rewards/keyword_inclusion_reward/std": 0.05127105638384819,
|
| 1326 |
+
"step": 260
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"clip_ratio/high_max": 0.0,
|
| 1330 |
+
"clip_ratio/high_mean": 0.0,
|
| 1331 |
+
"clip_ratio/low_mean": 0.0,
|
| 1332 |
+
"clip_ratio/low_min": 0.0,
|
| 1333 |
+
"clip_ratio/region_mean": 0.0,
|
| 1334 |
+
"completions/clipped_ratio": 0.975,
|
| 1335 |
+
"completions/max_length": 32.0,
|
| 1336 |
+
"completions/max_terminated_length": 18.0,
|
| 1337 |
+
"completions/mean_length": 31.9375,
|
| 1338 |
+
"completions/mean_terminated_length": 17.7,
|
| 1339 |
+
"completions/min_length": 30.2,
|
| 1340 |
+
"completions/min_terminated_length": 17.4,
|
| 1341 |
+
"epoch": 0.3428201811125485,
|
| 1342 |
+
"grad_norm": 0.21370354294776917,
|
| 1343 |
+
"kl": 0.3591796875,
|
| 1344 |
+
"learning_rate": 0.00018405666034956844,
|
| 1345 |
+
"loss": 0.0144,
|
| 1346 |
+
"num_tokens": 293805.0,
|
| 1347 |
+
"reward": 2.886680746078491,
|
| 1348 |
+
"reward_std": 0.10467648506164551,
|
| 1349 |
+
"rewards/keyword_inclusion_reward/mean": 0.325000011920929,
|
| 1350 |
+
"rewards/keyword_inclusion_reward/std": 0.039965872466564176,
|
| 1351 |
+
"step": 265
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"clip_ratio/high_max": 0.0,
|
| 1355 |
+
"clip_ratio/high_mean": 0.0,
|
| 1356 |
+
"clip_ratio/low_mean": 0.0,
|
| 1357 |
+
"clip_ratio/low_min": 0.0,
|
| 1358 |
+
"clip_ratio/region_mean": 0.0,
|
| 1359 |
+
"completions/clipped_ratio": 0.975,
|
| 1360 |
+
"completions/max_length": 32.0,
|
| 1361 |
+
"completions/max_terminated_length": 18.0,
|
| 1362 |
+
"completions/mean_length": 31.89375,
|
| 1363 |
+
"completions/mean_terminated_length": 17.2,
|
| 1364 |
+
"completions/min_length": 29.2,
|
| 1365 |
+
"completions/min_terminated_length": 16.4,
|
| 1366 |
+
"epoch": 0.34928848641655885,
|
| 1367 |
+
"grad_norm": 0.17620764672756195,
|
| 1368 |
+
"kl": 0.3935546875,
|
| 1369 |
+
"learning_rate": 0.00018309844692743283,
|
| 1370 |
+
"loss": 0.0157,
|
| 1371 |
+
"num_tokens": 299424.0,
|
| 1372 |
+
"reward": 2.9236894607543946,
|
| 1373 |
+
"reward_std": 0.052338242530822754,
|
| 1374 |
+
"rewards/keyword_inclusion_reward/mean": 0.3291666805744171,
|
| 1375 |
+
"rewards/keyword_inclusion_reward/std": 0.023570226132869722,
|
| 1376 |
+
"step": 270
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"clip_ratio/high_max": 0.0,
|
| 1380 |
+
"clip_ratio/high_mean": 0.0,
|
| 1381 |
+
"clip_ratio/low_mean": 0.0,
|
| 1382 |
+
"clip_ratio/low_min": 0.0,
|
| 1383 |
+
"clip_ratio/region_mean": 0.0,
|
| 1384 |
+
"completions/clipped_ratio": 0.96875,
|
| 1385 |
+
"completions/max_length": 32.0,
|
| 1386 |
+
"completions/max_terminated_length": 11.8,
|
| 1387 |
+
"completions/mean_length": 31.84375,
|
| 1388 |
+
"completions/mean_terminated_length": 10.833333587646484,
|
| 1389 |
+
"completions/min_length": 29.2,
|
| 1390 |
+
"completions/min_terminated_length": 10.0,
|
| 1391 |
+
"epoch": 0.35575679172056923,
|
| 1392 |
+
"grad_norm": 0.2770165205001831,
|
| 1393 |
+
"kl": 0.35625,
|
| 1394 |
+
"learning_rate": 0.00018211492091337042,
|
| 1395 |
+
"loss": 0.0143,
|
| 1396 |
+
"num_tokens": 305035.0,
|
| 1397 |
+
"reward": 2.7941588878631594,
|
| 1398 |
+
"reward_std": 0.18318384885787964,
|
| 1399 |
+
"rewards/keyword_inclusion_reward/mean": 0.3145833432674408,
|
| 1400 |
+
"rewards/keyword_inclusion_reward/std": 0.05612155273556709,
|
| 1401 |
+
"step": 275
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"clip_ratio/high_max": 0.0,
|
| 1405 |
+
"clip_ratio/high_mean": 0.0,
|
| 1406 |
+
"clip_ratio/low_mean": 0.0,
|
| 1407 |
+
"clip_ratio/low_min": 0.0,
|
| 1408 |
+
"clip_ratio/region_mean": 0.0,
|
| 1409 |
+
"completions/clipped_ratio": 0.95,
|
| 1410 |
+
"completions/max_length": 32.0,
|
| 1411 |
+
"completions/max_terminated_length": 22.6,
|
| 1412 |
+
"completions/mean_length": 31.66875,
|
| 1413 |
+
"completions/mean_terminated_length": 20.3,
|
| 1414 |
+
"completions/min_length": 24.4,
|
| 1415 |
+
"completions/min_terminated_length": 18.0,
|
| 1416 |
+
"epoch": 0.36222509702457956,
|
| 1417 |
+
"grad_norm": 0.13597851991653442,
|
| 1418 |
+
"kl": 0.3595703125,
|
| 1419 |
+
"learning_rate": 0.00018110638189893267,
|
| 1420 |
+
"loss": 0.0138,
|
| 1421 |
+
"num_tokens": 310616.0,
|
| 1422 |
+
"reward": 2.812663221359253,
|
| 1423 |
+
"reward_std": 0.20935297012329102,
|
| 1424 |
+
"rewards/keyword_inclusion_reward/mean": 0.31666667461395265,
|
| 1425 |
+
"rewards/keyword_inclusion_reward/std": 0.07281493991613389,
|
| 1426 |
+
"step": 280
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"clip_ratio/high_max": 0.0,
|
| 1430 |
+
"clip_ratio/high_mean": 0.0,
|
| 1431 |
+
"clip_ratio/low_mean": 0.0,
|
| 1432 |
+
"clip_ratio/low_min": 0.0,
|
| 1433 |
+
"clip_ratio/region_mean": 0.0,
|
| 1434 |
+
"completions/clipped_ratio": 0.93125,
|
| 1435 |
+
"completions/max_length": 32.0,
|
| 1436 |
+
"completions/max_terminated_length": 24.4,
|
| 1437 |
+
"completions/mean_length": 31.575,
|
| 1438 |
+
"completions/mean_terminated_length": 21.683333587646484,
|
| 1439 |
+
"completions/min_length": 23.0,
|
| 1440 |
+
"completions/min_terminated_length": 16.6,
|
| 1441 |
+
"epoch": 0.36869340232858994,
|
| 1442 |
+
"grad_norm": 0.13746124505996704,
|
| 1443 |
+
"kl": 0.3494140625,
|
| 1444 |
+
"learning_rate": 0.00018007313709487334,
|
| 1445 |
+
"loss": 0.014,
|
| 1446 |
+
"num_tokens": 316070.0,
|
| 1447 |
+
"reward": 2.9236894607543946,
|
| 1448 |
+
"reward_std": 0.052338242530822754,
|
| 1449 |
+
"rewards/keyword_inclusion_reward/mean": 0.3291666805744171,
|
| 1450 |
+
"rewards/keyword_inclusion_reward/std": 0.023570226132869722,
|
| 1451 |
+
"step": 285
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"clip_ratio/high_max": 0.0,
|
| 1455 |
+
"clip_ratio/high_mean": 0.0,
|
| 1456 |
+
"clip_ratio/low_mean": 0.0,
|
| 1457 |
+
"clip_ratio/low_min": 0.0,
|
| 1458 |
+
"clip_ratio/region_mean": 0.0,
|
| 1459 |
+
"completions/clipped_ratio": 0.95,
|
| 1460 |
+
"completions/max_length": 32.0,
|
| 1461 |
+
"completions/max_terminated_length": 23.2,
|
| 1462 |
+
"completions/mean_length": 31.39375,
|
| 1463 |
+
"completions/mean_terminated_length": 17.233333587646484,
|
| 1464 |
+
"completions/min_length": 17.8,
|
| 1465 |
+
"completions/min_terminated_length": 11.4,
|
| 1466 |
+
"epoch": 0.37516170763260026,
|
| 1467 |
+
"grad_norm": 0.42286813259124756,
|
| 1468 |
+
"kl": 0.3724609375,
|
| 1469 |
+
"learning_rate": 0.00017901550123756906,
|
| 1470 |
+
"loss": 0.0059,
|
| 1471 |
+
"num_tokens": 321569.0,
|
| 1472 |
+
"reward": 2.8311676502227785,
|
| 1473 |
+
"reward_std": 0.18318384885787964,
|
| 1474 |
+
"rewards/keyword_inclusion_reward/mean": 0.31875001192092894,
|
| 1475 |
+
"rewards/keyword_inclusion_reward/std": 0.050581476837396624,
|
| 1476 |
+
"step": 290
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"clip_ratio/high_max": 0.0,
|
| 1480 |
+
"clip_ratio/high_mean": 0.0,
|
| 1481 |
+
"clip_ratio/low_mean": 0.0,
|
| 1482 |
+
"clip_ratio/low_min": 0.0,
|
| 1483 |
+
"clip_ratio/region_mean": 0.0,
|
| 1484 |
+
"completions/clipped_ratio": 0.95,
|
| 1485 |
+
"completions/max_length": 32.0,
|
| 1486 |
+
"completions/max_terminated_length": 24.0,
|
| 1487 |
+
"completions/mean_length": 31.8625,
|
| 1488 |
+
"completions/mean_terminated_length": 23.333333587646486,
|
| 1489 |
+
"completions/min_length": 29.4,
|
| 1490 |
+
"completions/min_terminated_length": 23.0,
|
| 1491 |
+
"epoch": 0.3816300129366106,
|
| 1492 |
+
"grad_norm": 0.17354200780391693,
|
| 1493 |
+
"kl": 0.3912109375,
|
| 1494 |
+
"learning_rate": 0.00017793379649314744,
|
| 1495 |
+
"loss": 0.0156,
|
| 1496 |
+
"num_tokens": 327143.0,
|
| 1497 |
+
"reward": 2.8681763648986816,
|
| 1498 |
+
"reward_std": 0.13084560632705688,
|
| 1499 |
+
"rewards/keyword_inclusion_reward/mean": 0.3229166746139526,
|
| 1500 |
+
"rewards/keyword_inclusion_reward/std": 0.04457640573382378,
|
| 1501 |
+
"step": 295
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.38809831824062097,
|
| 1505 |
+
"grad_norm": 0.23563814163208008,
|
| 1506 |
+
"learning_rate": 0.00017682835235935236,
|
| 1507 |
+
"loss": 0.014,
|
| 1508 |
+
"step": 300
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.38809831824062097,
|
| 1512 |
+
"eval_clip_ratio/high_max": 0.0,
|
| 1513 |
+
"eval_clip_ratio/high_mean": 0.0,
|
| 1514 |
+
"eval_clip_ratio/low_mean": 0.0,
|
| 1515 |
+
"eval_clip_ratio/low_min": 0.0,
|
| 1516 |
+
"eval_clip_ratio/region_mean": 0.0,
|
| 1517 |
+
"eval_completions/clipped_ratio": 0.9840425531914894,
|
| 1518 |
+
"eval_completions/max_length": 32.0,
|
| 1519 |
+
"eval_completions/max_terminated_length": 7.085106382978723,
|
| 1520 |
+
"eval_completions/mean_length": 31.93218085106383,
|
| 1521 |
+
"eval_completions/mean_terminated_length": 7.085106382978723,
|
| 1522 |
+
"eval_completions/min_length": 30.914893617021278,
|
| 1523 |
+
"eval_completions/min_terminated_length": 7.085106382978723,
|
| 1524 |
+
"eval_kl": 0.3568816489361702,
|
| 1525 |
+
"eval_loss": 0.0143232811242342,
|
| 1526 |
+
"eval_num_tokens": 332728.0,
|
| 1527 |
+
"eval_reward": 2.7559690475463867,
|
| 1528 |
+
"eval_reward_std": 0.26725911079569065,
|
| 1529 |
+
"eval_rewards/keyword_inclusion_reward/mean": 0.31028369639782194,
|
| 1530 |
+
"eval_rewards/keyword_inclusion_reward/std": 0.0670770841076019,
|
| 1531 |
+
"eval_runtime": 77.8145,
|
| 1532 |
+
"eval_samples_per_second": 4.755,
|
| 1533 |
+
"eval_steps_per_second": 0.308,
|
| 1534 |
+
"step": 300
|
| 1535 |
}
|
| 1536 |
],
|
| 1537 |
"logging_steps": 5,
|
| 1538 |
"max_steps": 1000,
|
| 1539 |
+
"num_input_tokens_seen": 332728,
|
| 1540 |
"num_train_epochs": 2,
|
| 1541 |
"save_steps": 100,
|
| 1542 |
"stateful_callbacks": {
|