Training in progress, epoch 3, checkpoint
Browse files- last-checkpoint/global_step915/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step915/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step915/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step915/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step915/zero_pp_rank_0_mp_rank_00_model_states.pt +3 -0
- last-checkpoint/global_step915/zero_pp_rank_1_mp_rank_00_model_states.pt +3 -0
- last-checkpoint/global_step915/zero_pp_rank_2_mp_rank_00_model_states.pt +3 -0
- last-checkpoint/global_step915/zero_pp_rank_3_mp_rank_00_model_states.pt +3 -0
- last-checkpoint/latest +1 -1
- last-checkpoint/model-00001-of-00004.safetensors +1 -1
- last-checkpoint/model-00002-of-00004.safetensors +1 -1
- last-checkpoint/model-00003-of-00004.safetensors +1 -1
- last-checkpoint/model-00004-of-00004.safetensors +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 +469 -3
last-checkpoint/global_step915/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 sha256:8dc3914fa38b4e4c831957ec8d11121c6d7d99458de538b9a5297f3db0b48fc1
|
| 3 |
+
size 17075366956
|
last-checkpoint/global_step915/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:daeb306441a8e305d688d504d959f48bde9cb7b0481af4b123a46acde68d3be3
|
| 3 |
+
size 17075366956
|
last-checkpoint/global_step915/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 sha256:37a6b3fd50c04f9c583042eb5f2f54b8b1ecb13b945d799abb65ca70e44433b3
|
| 3 |
+
size 17075366956
|
last-checkpoint/global_step915/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:d7993b99fed9a9bf82f2fb8803ecf090c237f0a42e5e7fadcc1bc19b1004ee66
|
| 3 |
+
size 17075366956
|
last-checkpoint/global_step915/zero_pp_rank_0_mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20a7fd50be01694e49f9de3f8f1272257c8c03f4ad98ad21333c50f043c9f4e2
|
| 3 |
+
size 131892
|
last-checkpoint/global_step915/zero_pp_rank_1_mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bdc85537b53b2e8316afc05428c47945418ef5624b9e5a3e7fb0883dd92beafa
|
| 3 |
+
size 131828
|
last-checkpoint/global_step915/zero_pp_rank_2_mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a36623a59edfd11a4af114aed7acad913054b7c916187ceb13f66f370b5df031
|
| 3 |
+
size 131828
|
last-checkpoint/global_step915/zero_pp_rank_3_mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a971b41276ea7e04ae399c49141ed5b590cd146d7caaaf0c73534af09b8a01c1
|
| 3 |
+
size 131828
|
last-checkpoint/latest
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
|
|
|
|
| 1 |
+
global_step915
|
last-checkpoint/model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4995496656
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a948184af7a6b626a16e84585b352af277a0c889e58f35b3060a4d06b2d9c8d8
|
| 3 |
size 4995496656
|
last-checkpoint/model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4982953168
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e8498cf11f8a7d244236bff406a1516d3965c4bf8e94d45ec31aff6f4fb7851
|
| 3 |
size 4982953168
|
last-checkpoint/model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4982953200
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30f814437f099d7cf822fcecfadfb665bcb15c94551a4e27ea560b0983b10ef4
|
| 3 |
size 4982953200
|
last-checkpoint/model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2113988336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7fca10058d491211c29532dd0e5c876d9d4001ae759ed094a3d8e57dd3e214c0
|
| 3 |
size 2113988336
|
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 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab7cf0d34d60dfda516cf9661904550e2e294e723edd07c25c738f05e8ba92d1
|
| 3 |
size 15024
|
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 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7b183324e8227a51a9556d86b2ad893a8c4c52205ed4a737356c6611dac7353
|
| 3 |
size 15024
|
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 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac69e994090f4818cb1fa6f6cefa363178552c3c731c6507ff195bcb07fd5bef
|
| 3 |
size 15024
|
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 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f68e61b63402f8afb1f69c960f7944965655dac11e3ccf29919c282f23931f86
|
| 3 |
size 15024
|
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 sha256:89cc83db19eba05de12a39437c27831b7029da4dd2494fc1e52d31a689e731bb
|
| 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":
|
| 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,
|
|
@@ -969,6 +969,472 @@
|
|
| 969 |
"eval_samples_per_second": 13.141,
|
| 970 |
"eval_steps_per_second": 0.469,
|
| 971 |
"step": 610
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 972 |
}
|
| 973 |
],
|
| 974 |
"logging_steps": 10,
|
|
@@ -983,7 +1449,7 @@
|
|
| 983 |
"should_evaluate": false,
|
| 984 |
"should_log": false,
|
| 985 |
"should_save": true,
|
| 986 |
-
"should_training_stop":
|
| 987 |
},
|
| 988 |
"attributes": {}
|
| 989 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 915,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 969 |
"eval_samples_per_second": 13.141,
|
| 970 |
"eval_steps_per_second": 0.469,
|
| 971 |
"step": 610
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.0327868852459017,
|
| 975 |
+
"grad_norm": 1.3613376787035118e-19,
|
| 976 |
+
"learning_rate": 1.792223572296476e-07,
|
| 977 |
+
"logits/chosen": 268.0,
|
| 978 |
+
"logits/rejected": 274.0,
|
| 979 |
+
"logps/chosen": -740.0,
|
| 980 |
+
"logps/rejected": -1216.0,
|
| 981 |
+
"loss": 0.0,
|
| 982 |
+
"rewards/accuracies": 1.0,
|
| 983 |
+
"rewards/chosen": 120.5,
|
| 984 |
+
"rewards/margins": 155.0,
|
| 985 |
+
"rewards/rejected": -34.25,
|
| 986 |
+
"step": 620
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 2.0655737704918034,
|
| 990 |
+
"grad_norm": 0.0,
|
| 991 |
+
"learning_rate": 1.7314702308626974e-07,
|
| 992 |
+
"logits/chosen": 268.0,
|
| 993 |
+
"logits/rejected": 274.0,
|
| 994 |
+
"logps/chosen": -748.0,
|
| 995 |
+
"logps/rejected": -1144.0,
|
| 996 |
+
"loss": 0.0,
|
| 997 |
+
"rewards/accuracies": 1.0,
|
| 998 |
+
"rewards/chosen": 115.5,
|
| 999 |
+
"rewards/margins": 148.0,
|
| 1000 |
+
"rewards/rejected": -32.5,
|
| 1001 |
+
"step": 630
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 2.098360655737705,
|
| 1005 |
+
"grad_norm": 0.0,
|
| 1006 |
+
"learning_rate": 1.6707168894289186e-07,
|
| 1007 |
+
"logits/chosen": 268.0,
|
| 1008 |
+
"logits/rejected": 274.0,
|
| 1009 |
+
"logps/chosen": -816.0,
|
| 1010 |
+
"logps/rejected": -1152.0,
|
| 1011 |
+
"loss": 0.0,
|
| 1012 |
+
"rewards/accuracies": 1.0,
|
| 1013 |
+
"rewards/chosen": 112.0,
|
| 1014 |
+
"rewards/margins": 145.0,
|
| 1015 |
+
"rewards/rejected": -32.75,
|
| 1016 |
+
"step": 640
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.1311475409836067,
|
| 1020 |
+
"grad_norm": 0.0,
|
| 1021 |
+
"learning_rate": 1.6099635479951396e-07,
|
| 1022 |
+
"logits/chosen": 268.0,
|
| 1023 |
+
"logits/rejected": 274.0,
|
| 1024 |
+
"logps/chosen": -792.0,
|
| 1025 |
+
"logps/rejected": -1208.0,
|
| 1026 |
+
"loss": 0.0,
|
| 1027 |
+
"rewards/accuracies": 1.0,
|
| 1028 |
+
"rewards/chosen": 114.0,
|
| 1029 |
+
"rewards/margins": 148.0,
|
| 1030 |
+
"rewards/rejected": -34.5,
|
| 1031 |
+
"step": 650
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.1639344262295084,
|
| 1035 |
+
"grad_norm": 6.520051993303297e-20,
|
| 1036 |
+
"learning_rate": 1.5492102065613608e-07,
|
| 1037 |
+
"logits/chosen": 268.0,
|
| 1038 |
+
"logits/rejected": 274.0,
|
| 1039 |
+
"logps/chosen": -744.0,
|
| 1040 |
+
"logps/rejected": -1160.0,
|
| 1041 |
+
"loss": 0.0,
|
| 1042 |
+
"rewards/accuracies": 1.0,
|
| 1043 |
+
"rewards/chosen": 111.5,
|
| 1044 |
+
"rewards/margins": 145.0,
|
| 1045 |
+
"rewards/rejected": -33.0,
|
| 1046 |
+
"step": 660
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 2.19672131147541,
|
| 1050 |
+
"grad_norm": 0.0,
|
| 1051 |
+
"learning_rate": 1.488456865127582e-07,
|
| 1052 |
+
"logits/chosen": 268.0,
|
| 1053 |
+
"logits/rejected": 274.0,
|
| 1054 |
+
"logps/chosen": -808.0,
|
| 1055 |
+
"logps/rejected": -1248.0,
|
| 1056 |
+
"loss": 0.0,
|
| 1057 |
+
"rewards/accuracies": 1.0,
|
| 1058 |
+
"rewards/chosen": 115.0,
|
| 1059 |
+
"rewards/margins": 150.0,
|
| 1060 |
+
"rewards/rejected": -35.0,
|
| 1061 |
+
"step": 670
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 2.2295081967213113,
|
| 1065 |
+
"grad_norm": 8.585906103936093e-32,
|
| 1066 |
+
"learning_rate": 1.427703523693803e-07,
|
| 1067 |
+
"logits/chosen": 268.0,
|
| 1068 |
+
"logits/rejected": 274.0,
|
| 1069 |
+
"logps/chosen": -776.0,
|
| 1070 |
+
"logps/rejected": -1144.0,
|
| 1071 |
+
"loss": 0.0,
|
| 1072 |
+
"rewards/accuracies": 1.0,
|
| 1073 |
+
"rewards/chosen": 119.0,
|
| 1074 |
+
"rewards/margins": 152.0,
|
| 1075 |
+
"rewards/rejected": -33.25,
|
| 1076 |
+
"step": 680
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 2.262295081967213,
|
| 1080 |
+
"grad_norm": 1.1432686013314791e-21,
|
| 1081 |
+
"learning_rate": 1.3669501822600243e-07,
|
| 1082 |
+
"logits/chosen": 268.0,
|
| 1083 |
+
"logits/rejected": 274.0,
|
| 1084 |
+
"logps/chosen": -796.0,
|
| 1085 |
+
"logps/rejected": -1224.0,
|
| 1086 |
+
"loss": 0.0,
|
| 1087 |
+
"rewards/accuracies": 1.0,
|
| 1088 |
+
"rewards/chosen": 111.0,
|
| 1089 |
+
"rewards/margins": 146.0,
|
| 1090 |
+
"rewards/rejected": -34.75,
|
| 1091 |
+
"step": 690
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 2.2950819672131146,
|
| 1095 |
+
"grad_norm": 3.757902374283223e-21,
|
| 1096 |
+
"learning_rate": 1.3061968408262452e-07,
|
| 1097 |
+
"logits/chosen": 268.0,
|
| 1098 |
+
"logits/rejected": 274.0,
|
| 1099 |
+
"logps/chosen": -776.0,
|
| 1100 |
+
"logps/rejected": -1184.0,
|
| 1101 |
+
"loss": 0.0,
|
| 1102 |
+
"rewards/accuracies": 1.0,
|
| 1103 |
+
"rewards/chosen": 114.0,
|
| 1104 |
+
"rewards/margins": 146.0,
|
| 1105 |
+
"rewards/rejected": -32.5,
|
| 1106 |
+
"step": 700
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 2.3278688524590163,
|
| 1110 |
+
"grad_norm": 0.0,
|
| 1111 |
+
"learning_rate": 1.2454434993924665e-07,
|
| 1112 |
+
"logits/chosen": 268.0,
|
| 1113 |
+
"logits/rejected": 274.0,
|
| 1114 |
+
"logps/chosen": -788.0,
|
| 1115 |
+
"logps/rejected": -1216.0,
|
| 1116 |
+
"loss": 0.0,
|
| 1117 |
+
"rewards/accuracies": 1.0,
|
| 1118 |
+
"rewards/chosen": 114.0,
|
| 1119 |
+
"rewards/margins": 149.0,
|
| 1120 |
+
"rewards/rejected": -35.0,
|
| 1121 |
+
"step": 710
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 2.360655737704918,
|
| 1125 |
+
"grad_norm": 5.760183258482964e-35,
|
| 1126 |
+
"learning_rate": 1.1846901579586877e-07,
|
| 1127 |
+
"logits/chosen": 268.0,
|
| 1128 |
+
"logits/rejected": 274.0,
|
| 1129 |
+
"logps/chosen": -812.0,
|
| 1130 |
+
"logps/rejected": -1160.0,
|
| 1131 |
+
"loss": 0.0,
|
| 1132 |
+
"rewards/accuracies": 1.0,
|
| 1133 |
+
"rewards/chosen": 114.0,
|
| 1134 |
+
"rewards/margins": 146.0,
|
| 1135 |
+
"rewards/rejected": -32.5,
|
| 1136 |
+
"step": 720
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.3934426229508197,
|
| 1140 |
+
"grad_norm": 3.825327885849806e-35,
|
| 1141 |
+
"learning_rate": 1.1239368165249088e-07,
|
| 1142 |
+
"logits/chosen": 268.0,
|
| 1143 |
+
"logits/rejected": 274.0,
|
| 1144 |
+
"logps/chosen": -752.0,
|
| 1145 |
+
"logps/rejected": -1208.0,
|
| 1146 |
+
"loss": 0.0,
|
| 1147 |
+
"rewards/accuracies": 1.0,
|
| 1148 |
+
"rewards/chosen": 115.0,
|
| 1149 |
+
"rewards/margins": 150.0,
|
| 1150 |
+
"rewards/rejected": -34.5,
|
| 1151 |
+
"step": 730
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 2.4262295081967213,
|
| 1155 |
+
"grad_norm": 9.742010576487992e-17,
|
| 1156 |
+
"learning_rate": 1.0631834750911299e-07,
|
| 1157 |
+
"logits/chosen": 268.0,
|
| 1158 |
+
"logits/rejected": 274.0,
|
| 1159 |
+
"logps/chosen": -720.0,
|
| 1160 |
+
"logps/rejected": -1224.0,
|
| 1161 |
+
"loss": 0.0,
|
| 1162 |
+
"rewards/accuracies": 1.0,
|
| 1163 |
+
"rewards/chosen": 114.0,
|
| 1164 |
+
"rewards/margins": 148.0,
|
| 1165 |
+
"rewards/rejected": -34.5,
|
| 1166 |
+
"step": 740
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 2.459016393442623,
|
| 1170 |
+
"grad_norm": 1.3910608136580302e-30,
|
| 1171 |
+
"learning_rate": 1.0024301336573512e-07,
|
| 1172 |
+
"logits/chosen": 268.0,
|
| 1173 |
+
"logits/rejected": 274.0,
|
| 1174 |
+
"logps/chosen": -792.0,
|
| 1175 |
+
"logps/rejected": -1160.0,
|
| 1176 |
+
"loss": 0.0,
|
| 1177 |
+
"rewards/accuracies": 1.0,
|
| 1178 |
+
"rewards/chosen": 115.5,
|
| 1179 |
+
"rewards/margins": 149.0,
|
| 1180 |
+
"rewards/rejected": -33.5,
|
| 1181 |
+
"step": 750
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 2.4918032786885247,
|
| 1185 |
+
"grad_norm": 5.4809388850402516e-21,
|
| 1186 |
+
"learning_rate": 9.416767922235723e-08,
|
| 1187 |
+
"logits/chosen": 268.0,
|
| 1188 |
+
"logits/rejected": 274.0,
|
| 1189 |
+
"logps/chosen": -736.0,
|
| 1190 |
+
"logps/rejected": -1184.0,
|
| 1191 |
+
"loss": 0.0,
|
| 1192 |
+
"rewards/accuracies": 1.0,
|
| 1193 |
+
"rewards/chosen": 119.5,
|
| 1194 |
+
"rewards/margins": 152.0,
|
| 1195 |
+
"rewards/rejected": -32.75,
|
| 1196 |
+
"step": 760
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 2.5245901639344264,
|
| 1200 |
+
"grad_norm": 5.540153547398696e-15,
|
| 1201 |
+
"learning_rate": 8.809234507897934e-08,
|
| 1202 |
+
"logits/chosen": 268.0,
|
| 1203 |
+
"logits/rejected": 274.0,
|
| 1204 |
+
"logps/chosen": -784.0,
|
| 1205 |
+
"logps/rejected": -1144.0,
|
| 1206 |
+
"loss": 0.0,
|
| 1207 |
+
"rewards/accuracies": 1.0,
|
| 1208 |
+
"rewards/chosen": 116.5,
|
| 1209 |
+
"rewards/margins": 150.0,
|
| 1210 |
+
"rewards/rejected": -33.25,
|
| 1211 |
+
"step": 770
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 2.557377049180328,
|
| 1215 |
+
"grad_norm": 0.0,
|
| 1216 |
+
"learning_rate": 8.201701093560146e-08,
|
| 1217 |
+
"logits/chosen": 268.0,
|
| 1218 |
+
"logits/rejected": 274.0,
|
| 1219 |
+
"logps/chosen": -776.0,
|
| 1220 |
+
"logps/rejected": -1168.0,
|
| 1221 |
+
"loss": 0.0,
|
| 1222 |
+
"rewards/accuracies": 1.0,
|
| 1223 |
+
"rewards/chosen": 117.0,
|
| 1224 |
+
"rewards/margins": 150.0,
|
| 1225 |
+
"rewards/rejected": -33.0,
|
| 1226 |
+
"step": 780
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 2.5901639344262293,
|
| 1230 |
+
"grad_norm": 0.0,
|
| 1231 |
+
"learning_rate": 7.594167679222357e-08,
|
| 1232 |
+
"logits/chosen": 268.0,
|
| 1233 |
+
"logits/rejected": 274.0,
|
| 1234 |
+
"logps/chosen": -768.0,
|
| 1235 |
+
"logps/rejected": -1216.0,
|
| 1236 |
+
"loss": 0.0,
|
| 1237 |
+
"rewards/accuracies": 1.0,
|
| 1238 |
+
"rewards/chosen": 116.5,
|
| 1239 |
+
"rewards/margins": 151.0,
|
| 1240 |
+
"rewards/rejected": -34.25,
|
| 1241 |
+
"step": 790
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.6229508196721314,
|
| 1245 |
+
"grad_norm": 3.694790735744769e-15,
|
| 1246 |
+
"learning_rate": 6.986634264884568e-08,
|
| 1247 |
+
"logits/chosen": 268.0,
|
| 1248 |
+
"logits/rejected": 274.0,
|
| 1249 |
+
"logps/chosen": -796.0,
|
| 1250 |
+
"logps/rejected": -1152.0,
|
| 1251 |
+
"loss": 0.0,
|
| 1252 |
+
"rewards/accuracies": 1.0,
|
| 1253 |
+
"rewards/chosen": 114.0,
|
| 1254 |
+
"rewards/margins": 147.0,
|
| 1255 |
+
"rewards/rejected": -33.0,
|
| 1256 |
+
"step": 800
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.6557377049180326,
|
| 1260 |
+
"grad_norm": 8.419354262347816e-13,
|
| 1261 |
+
"learning_rate": 6.37910085054678e-08,
|
| 1262 |
+
"logits/chosen": 268.0,
|
| 1263 |
+
"logits/rejected": 272.0,
|
| 1264 |
+
"logps/chosen": -780.0,
|
| 1265 |
+
"logps/rejected": -1160.0,
|
| 1266 |
+
"loss": 0.0,
|
| 1267 |
+
"rewards/accuracies": 1.0,
|
| 1268 |
+
"rewards/chosen": 117.5,
|
| 1269 |
+
"rewards/margins": 150.0,
|
| 1270 |
+
"rewards/rejected": -33.0,
|
| 1271 |
+
"step": 810
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 2.6885245901639343,
|
| 1275 |
+
"grad_norm": 3.7751524716356925e-16,
|
| 1276 |
+
"learning_rate": 5.771567436208991e-08,
|
| 1277 |
+
"logits/chosen": 268.0,
|
| 1278 |
+
"logits/rejected": 274.0,
|
| 1279 |
+
"logps/chosen": -760.0,
|
| 1280 |
+
"logps/rejected": -1184.0,
|
| 1281 |
+
"loss": 0.0,
|
| 1282 |
+
"rewards/accuracies": 1.0,
|
| 1283 |
+
"rewards/chosen": 115.0,
|
| 1284 |
+
"rewards/margins": 148.0,
|
| 1285 |
+
"rewards/rejected": -33.0,
|
| 1286 |
+
"step": 820
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 2.721311475409836,
|
| 1290 |
+
"grad_norm": 1.2321456713468952e-20,
|
| 1291 |
+
"learning_rate": 5.164034021871203e-08,
|
| 1292 |
+
"logits/chosen": 268.0,
|
| 1293 |
+
"logits/rejected": 274.0,
|
| 1294 |
+
"logps/chosen": -820.0,
|
| 1295 |
+
"logps/rejected": -1168.0,
|
| 1296 |
+
"loss": 0.0,
|
| 1297 |
+
"rewards/accuracies": 1.0,
|
| 1298 |
+
"rewards/chosen": 112.5,
|
| 1299 |
+
"rewards/margins": 146.0,
|
| 1300 |
+
"rewards/rejected": -32.75,
|
| 1301 |
+
"step": 830
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 2.7540983606557377,
|
| 1305 |
+
"grad_norm": 2.831509858410855e-15,
|
| 1306 |
+
"learning_rate": 4.5565006075334144e-08,
|
| 1307 |
+
"logits/chosen": 268.0,
|
| 1308 |
+
"logits/rejected": 274.0,
|
| 1309 |
+
"logps/chosen": -744.0,
|
| 1310 |
+
"logps/rejected": -1184.0,
|
| 1311 |
+
"loss": 0.0,
|
| 1312 |
+
"rewards/accuracies": 1.0,
|
| 1313 |
+
"rewards/chosen": 116.5,
|
| 1314 |
+
"rewards/margins": 150.0,
|
| 1315 |
+
"rewards/rejected": -32.75,
|
| 1316 |
+
"step": 840
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 2.7868852459016393,
|
| 1320 |
+
"grad_norm": 7.649524846493369e-22,
|
| 1321 |
+
"learning_rate": 3.9489671931956255e-08,
|
| 1322 |
+
"logits/chosen": 268.0,
|
| 1323 |
+
"logits/rejected": 274.0,
|
| 1324 |
+
"logps/chosen": -808.0,
|
| 1325 |
+
"logps/rejected": -1192.0,
|
| 1326 |
+
"loss": 0.0,
|
| 1327 |
+
"rewards/accuracies": 1.0,
|
| 1328 |
+
"rewards/chosen": 110.0,
|
| 1329 |
+
"rewards/margins": 144.0,
|
| 1330 |
+
"rewards/rejected": -34.25,
|
| 1331 |
+
"step": 850
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 2.819672131147541,
|
| 1335 |
+
"grad_norm": 0.0,
|
| 1336 |
+
"learning_rate": 3.341433778857837e-08,
|
| 1337 |
+
"logits/chosen": 268.0,
|
| 1338 |
+
"logits/rejected": 274.0,
|
| 1339 |
+
"logps/chosen": -784.0,
|
| 1340 |
+
"logps/rejected": -1192.0,
|
| 1341 |
+
"loss": 0.0,
|
| 1342 |
+
"rewards/accuracies": 1.0,
|
| 1343 |
+
"rewards/chosen": 117.5,
|
| 1344 |
+
"rewards/margins": 151.0,
|
| 1345 |
+
"rewards/rejected": -33.75,
|
| 1346 |
+
"step": 860
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.8524590163934427,
|
| 1350 |
+
"grad_norm": 0.0,
|
| 1351 |
+
"learning_rate": 2.7339003645200486e-08,
|
| 1352 |
+
"logits/chosen": 268.0,
|
| 1353 |
+
"logits/rejected": 274.0,
|
| 1354 |
+
"logps/chosen": -772.0,
|
| 1355 |
+
"logps/rejected": -1192.0,
|
| 1356 |
+
"loss": 0.0,
|
| 1357 |
+
"rewards/accuracies": 1.0,
|
| 1358 |
+
"rewards/chosen": 117.0,
|
| 1359 |
+
"rewards/margins": 150.0,
|
| 1360 |
+
"rewards/rejected": -33.5,
|
| 1361 |
+
"step": 870
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 2.8852459016393444,
|
| 1365 |
+
"grad_norm": 0.0,
|
| 1366 |
+
"learning_rate": 2.12636695018226e-08,
|
| 1367 |
+
"logits/chosen": 268.0,
|
| 1368 |
+
"logits/rejected": 274.0,
|
| 1369 |
+
"logps/chosen": -772.0,
|
| 1370 |
+
"logps/rejected": -1160.0,
|
| 1371 |
+
"loss": 0.0,
|
| 1372 |
+
"rewards/accuracies": 1.0,
|
| 1373 |
+
"rewards/chosen": 117.5,
|
| 1374 |
+
"rewards/margins": 150.0,
|
| 1375 |
+
"rewards/rejected": -32.75,
|
| 1376 |
+
"step": 880
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.918032786885246,
|
| 1380 |
+
"grad_norm": 2.740412262307959e-33,
|
| 1381 |
+
"learning_rate": 1.5188335358444714e-08,
|
| 1382 |
+
"logits/chosen": 268.0,
|
| 1383 |
+
"logits/rejected": 274.0,
|
| 1384 |
+
"logps/chosen": -752.0,
|
| 1385 |
+
"logps/rejected": -1232.0,
|
| 1386 |
+
"loss": 0.0,
|
| 1387 |
+
"rewards/accuracies": 1.0,
|
| 1388 |
+
"rewards/chosen": 121.5,
|
| 1389 |
+
"rewards/margins": 156.0,
|
| 1390 |
+
"rewards/rejected": -34.75,
|
| 1391 |
+
"step": 890
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 2.9508196721311473,
|
| 1395 |
+
"grad_norm": 4.338902475611692e-21,
|
| 1396 |
+
"learning_rate": 9.113001215066828e-09,
|
| 1397 |
+
"logits/chosen": 268.0,
|
| 1398 |
+
"logits/rejected": 274.0,
|
| 1399 |
+
"logps/chosen": -768.0,
|
| 1400 |
+
"logps/rejected": -1184.0,
|
| 1401 |
+
"loss": 0.0,
|
| 1402 |
+
"rewards/accuracies": 1.0,
|
| 1403 |
+
"rewards/chosen": 119.5,
|
| 1404 |
+
"rewards/margins": 153.0,
|
| 1405 |
+
"rewards/rejected": -33.5,
|
| 1406 |
+
"step": 900
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.9836065573770494,
|
| 1410 |
+
"grad_norm": 7.056451354101546e-30,
|
| 1411 |
+
"learning_rate": 3.0376670716889426e-09,
|
| 1412 |
+
"logits/chosen": 268.0,
|
| 1413 |
+
"logits/rejected": 274.0,
|
| 1414 |
+
"logps/chosen": -780.0,
|
| 1415 |
+
"logps/rejected": -1240.0,
|
| 1416 |
+
"loss": 0.0,
|
| 1417 |
+
"rewards/accuracies": 1.0,
|
| 1418 |
+
"rewards/chosen": 114.0,
|
| 1419 |
+
"rewards/margins": 149.0,
|
| 1420 |
+
"rewards/rejected": -35.5,
|
| 1421 |
+
"step": 910
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 3.0,
|
| 1425 |
+
"eval_logits/chosen": 268.0,
|
| 1426 |
+
"eval_logits/rejected": 274.0,
|
| 1427 |
+
"eval_logps/chosen": -796.0,
|
| 1428 |
+
"eval_logps/rejected": -1216.0,
|
| 1429 |
+
"eval_loss": 1.614817112492728e-12,
|
| 1430 |
+
"eval_rewards/accuracies": 1.0,
|
| 1431 |
+
"eval_rewards/chosen": 116.5,
|
| 1432 |
+
"eval_rewards/margins": 151.0,
|
| 1433 |
+
"eval_rewards/rejected": -34.75,
|
| 1434 |
+
"eval_runtime": 17.8148,
|
| 1435 |
+
"eval_samples_per_second": 11.002,
|
| 1436 |
+
"eval_steps_per_second": 0.393,
|
| 1437 |
+
"step": 915
|
| 1438 |
}
|
| 1439 |
],
|
| 1440 |
"logging_steps": 10,
|
|
|
|
| 1449 |
"should_evaluate": false,
|
| 1450 |
"should_log": false,
|
| 1451 |
"should_save": true,
|
| 1452 |
+
"should_training_stop": true
|
| 1453 |
},
|
| 1454 |
"attributes": {}
|
| 1455 |
}
|