Training in progress, step 420, 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 80013120
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b543d9e019f6d5c3cd652914901b2739520d85b9e6044fe4d75f753c8dd4dc9
|
| 3 |
size 80013120
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 41120084
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14c61ad2090140d21c6df9e0a61cd6d3225a4e43d63a4283da89db183775f6ae
|
| 3 |
size 41120084
|
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:c08654f8daac8b1091c235d2fb6bd8b249208c723b2dd501bc93b7ca776f4cba
|
| 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 sha256:24b34984058cd5169df3c13d6905d5f65a7c10a7cf4235e831bb570e73473147
|
| 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": 140,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1983,6 +1983,994 @@
|
|
| 1983 |
"eval_samples_per_second": 16.888,
|
| 1984 |
"eval_steps_per_second": 8.444,
|
| 1985 |
"step": 280
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1986 |
}
|
| 1987 |
],
|
| 1988 |
"logging_steps": 1,
|
|
@@ -2002,7 +2990,7 @@
|
|
| 2002 |
"attributes": {}
|
| 2003 |
}
|
| 2004 |
},
|
| 2005 |
-
"total_flos":
|
| 2006 |
"train_batch_size": 2,
|
| 2007 |
"trial_name": null,
|
| 2008 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.75,
|
| 5 |
"eval_steps": 140,
|
| 6 |
+
"global_step": 420,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1983 |
"eval_samples_per_second": 16.888,
|
| 1984 |
"eval_steps_per_second": 8.444,
|
| 1985 |
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 0.5017857142857143,
|
| 1989 |
+
"grad_norm": 0.3074837923049927,
|
| 1990 |
+
"learning_rate": 0.00010228459587429497,
|
| 1991 |
+
"loss": 1.1389,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.5035714285714286,
|
| 1996 |
+
"grad_norm": 0.3143484592437744,
|
| 1997 |
+
"learning_rate": 0.00010171351213038993,
|
| 1998 |
+
"loss": 0.9542,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.5053571428571428,
|
| 2003 |
+
"grad_norm": 0.3524804413318634,
|
| 2004 |
+
"learning_rate": 0.00010114237248023404,
|
| 2005 |
+
"loss": 0.8578,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.5071428571428571,
|
| 2010 |
+
"grad_norm": 0.335183322429657,
|
| 2011 |
+
"learning_rate": 0.00010057119555823085,
|
| 2012 |
+
"loss": 0.9228,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 0.5089285714285714,
|
| 2017 |
+
"grad_norm": 0.35537081956863403,
|
| 2018 |
+
"learning_rate": 0.0001,
|
| 2019 |
+
"loss": 0.9541,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.5107142857142857,
|
| 2024 |
+
"grad_norm": 0.3294563591480255,
|
| 2025 |
+
"learning_rate": 9.942880444176918e-05,
|
| 2026 |
+
"loss": 1.4223,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.5125,
|
| 2031 |
+
"grad_norm": 0.32628077268600464,
|
| 2032 |
+
"learning_rate": 9.8857627519766e-05,
|
| 2033 |
+
"loss": 1.2942,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.5142857142857142,
|
| 2038 |
+
"grad_norm": 0.32195404171943665,
|
| 2039 |
+
"learning_rate": 9.828648786961008e-05,
|
| 2040 |
+
"loss": 1.4239,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.5160714285714286,
|
| 2045 |
+
"grad_norm": 0.3283804655075073,
|
| 2046 |
+
"learning_rate": 9.771540412570504e-05,
|
| 2047 |
+
"loss": 1.0303,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.5178571428571429,
|
| 2052 |
+
"grad_norm": 0.356052428483963,
|
| 2053 |
+
"learning_rate": 9.71443949206304e-05,
|
| 2054 |
+
"loss": 1.0199,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 0.5196428571428572,
|
| 2059 |
+
"grad_norm": 0.39479124546051025,
|
| 2060 |
+
"learning_rate": 9.657347888453367e-05,
|
| 2061 |
+
"loss": 1.1343,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.5214285714285715,
|
| 2066 |
+
"grad_norm": 0.34791451692581177,
|
| 2067 |
+
"learning_rate": 9.60026746445227e-05,
|
| 2068 |
+
"loss": 1.521,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 0.5232142857142857,
|
| 2073 |
+
"grad_norm": 0.3614530861377716,
|
| 2074 |
+
"learning_rate": 9.543200082405768e-05,
|
| 2075 |
+
"loss": 1.2346,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 0.525,
|
| 2080 |
+
"grad_norm": 0.36958596110343933,
|
| 2081 |
+
"learning_rate": 9.486147604234371e-05,
|
| 2082 |
+
"loss": 1.0457,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 0.5267857142857143,
|
| 2087 |
+
"grad_norm": 0.4293418824672699,
|
| 2088 |
+
"learning_rate": 9.42911189137232e-05,
|
| 2089 |
+
"loss": 1.1071,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.5285714285714286,
|
| 2094 |
+
"grad_norm": 0.40408602356910706,
|
| 2095 |
+
"learning_rate": 9.372094804706867e-05,
|
| 2096 |
+
"loss": 1.3805,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 0.5303571428571429,
|
| 2101 |
+
"grad_norm": 0.3892784118652344,
|
| 2102 |
+
"learning_rate": 9.315098204517543e-05,
|
| 2103 |
+
"loss": 0.9136,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 0.5321428571428571,
|
| 2108 |
+
"grad_norm": 0.4003988206386566,
|
| 2109 |
+
"learning_rate": 9.258123950415479e-05,
|
| 2110 |
+
"loss": 1.3684,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.5339285714285714,
|
| 2115 |
+
"grad_norm": 0.37116503715515137,
|
| 2116 |
+
"learning_rate": 9.201173901282724e-05,
|
| 2117 |
+
"loss": 1.7824,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 0.5357142857142857,
|
| 2122 |
+
"grad_norm": 0.5286569595336914,
|
| 2123 |
+
"learning_rate": 9.144249915211605e-05,
|
| 2124 |
+
"loss": 1.9392,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 0.5375,
|
| 2129 |
+
"grad_norm": 0.18338526785373688,
|
| 2130 |
+
"learning_rate": 9.087353849444085e-05,
|
| 2131 |
+
"loss": 1.4422,
|
| 2132 |
+
"step": 301
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 0.5392857142857143,
|
| 2136 |
+
"grad_norm": 0.20191389322280884,
|
| 2137 |
+
"learning_rate": 9.030487560311186e-05,
|
| 2138 |
+
"loss": 1.5443,
|
| 2139 |
+
"step": 302
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 0.5410714285714285,
|
| 2143 |
+
"grad_norm": 0.19955602288246155,
|
| 2144 |
+
"learning_rate": 8.973652903172423e-05,
|
| 2145 |
+
"loss": 1.6521,
|
| 2146 |
+
"step": 303
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 0.5428571428571428,
|
| 2150 |
+
"grad_norm": 0.2253991812467575,
|
| 2151 |
+
"learning_rate": 8.916851732355255e-05,
|
| 2152 |
+
"loss": 1.6596,
|
| 2153 |
+
"step": 304
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 0.5446428571428571,
|
| 2157 |
+
"grad_norm": 0.18380412459373474,
|
| 2158 |
+
"learning_rate": 8.860085901094595e-05,
|
| 2159 |
+
"loss": 1.5462,
|
| 2160 |
+
"step": 305
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 0.5464285714285714,
|
| 2164 |
+
"grad_norm": 0.21318556368350983,
|
| 2165 |
+
"learning_rate": 8.803357261472343e-05,
|
| 2166 |
+
"loss": 1.6713,
|
| 2167 |
+
"step": 306
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 0.5482142857142858,
|
| 2171 |
+
"grad_norm": 0.20480670034885406,
|
| 2172 |
+
"learning_rate": 8.746667664356956e-05,
|
| 2173 |
+
"loss": 1.7537,
|
| 2174 |
+
"step": 307
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 0.55,
|
| 2178 |
+
"grad_norm": 0.21598470211029053,
|
| 2179 |
+
"learning_rate": 8.690018959343072e-05,
|
| 2180 |
+
"loss": 1.6955,
|
| 2181 |
+
"step": 308
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 0.5517857142857143,
|
| 2185 |
+
"grad_norm": 0.21172207593917847,
|
| 2186 |
+
"learning_rate": 8.633412994691144e-05,
|
| 2187 |
+
"loss": 1.7187,
|
| 2188 |
+
"step": 309
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 0.5535714285714286,
|
| 2192 |
+
"grad_norm": 0.22086191177368164,
|
| 2193 |
+
"learning_rate": 8.57685161726715e-05,
|
| 2194 |
+
"loss": 1.7579,
|
| 2195 |
+
"step": 310
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 0.5553571428571429,
|
| 2199 |
+
"grad_norm": 0.2266392558813095,
|
| 2200 |
+
"learning_rate": 8.520336672482338e-05,
|
| 2201 |
+
"loss": 1.7486,
|
| 2202 |
+
"step": 311
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 0.5571428571428572,
|
| 2206 |
+
"grad_norm": 0.2660948634147644,
|
| 2207 |
+
"learning_rate": 8.463870004233008e-05,
|
| 2208 |
+
"loss": 1.7903,
|
| 2209 |
+
"step": 312
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 0.5589285714285714,
|
| 2213 |
+
"grad_norm": 0.2297395020723343,
|
| 2214 |
+
"learning_rate": 8.407453454840357e-05,
|
| 2215 |
+
"loss": 1.8017,
|
| 2216 |
+
"step": 313
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 0.5607142857142857,
|
| 2220 |
+
"grad_norm": 0.21526682376861572,
|
| 2221 |
+
"learning_rate": 8.351088864990368e-05,
|
| 2222 |
+
"loss": 1.855,
|
| 2223 |
+
"step": 314
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 0.5625,
|
| 2227 |
+
"grad_norm": 0.2516147196292877,
|
| 2228 |
+
"learning_rate": 8.294778073673762e-05,
|
| 2229 |
+
"loss": 1.7103,
|
| 2230 |
+
"step": 315
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 0.5642857142857143,
|
| 2234 |
+
"grad_norm": 0.25641700625419617,
|
| 2235 |
+
"learning_rate": 8.238522918125983e-05,
|
| 2236 |
+
"loss": 1.9301,
|
| 2237 |
+
"step": 316
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 0.5660714285714286,
|
| 2241 |
+
"grad_norm": 0.26828062534332275,
|
| 2242 |
+
"learning_rate": 8.182325233767267e-05,
|
| 2243 |
+
"loss": 1.8575,
|
| 2244 |
+
"step": 317
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 0.5678571428571428,
|
| 2248 |
+
"grad_norm": 0.24787884950637817,
|
| 2249 |
+
"learning_rate": 8.126186854142752e-05,
|
| 2250 |
+
"loss": 2.0228,
|
| 2251 |
+
"step": 318
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 0.5696428571428571,
|
| 2255 |
+
"grad_norm": 0.23658955097198486,
|
| 2256 |
+
"learning_rate": 8.070109610862668e-05,
|
| 2257 |
+
"loss": 1.7813,
|
| 2258 |
+
"step": 319
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 0.5714285714285714,
|
| 2262 |
+
"grad_norm": 0.2818485498428345,
|
| 2263 |
+
"learning_rate": 8.014095333542548e-05,
|
| 2264 |
+
"loss": 1.7571,
|
| 2265 |
+
"step": 320
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 0.5732142857142857,
|
| 2269 |
+
"grad_norm": 0.24982373416423798,
|
| 2270 |
+
"learning_rate": 7.958145849743569e-05,
|
| 2271 |
+
"loss": 1.602,
|
| 2272 |
+
"step": 321
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 0.575,
|
| 2276 |
+
"grad_norm": 0.2815864682197571,
|
| 2277 |
+
"learning_rate": 7.902262984912909e-05,
|
| 2278 |
+
"loss": 1.7216,
|
| 2279 |
+
"step": 322
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 0.5767857142857142,
|
| 2283 |
+
"grad_norm": 0.2675464451313019,
|
| 2284 |
+
"learning_rate": 7.846448562324183e-05,
|
| 2285 |
+
"loss": 1.0704,
|
| 2286 |
+
"step": 323
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 0.5785714285714286,
|
| 2290 |
+
"grad_norm": 0.23938840627670288,
|
| 2291 |
+
"learning_rate": 7.79070440301796e-05,
|
| 2292 |
+
"loss": 1.282,
|
| 2293 |
+
"step": 324
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 0.5803571428571429,
|
| 2297 |
+
"grad_norm": 0.2834213972091675,
|
| 2298 |
+
"learning_rate": 7.735032325742355e-05,
|
| 2299 |
+
"loss": 1.7934,
|
| 2300 |
+
"step": 325
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 0.5821428571428572,
|
| 2304 |
+
"grad_norm": 0.3555513918399811,
|
| 2305 |
+
"learning_rate": 7.679434146893685e-05,
|
| 2306 |
+
"loss": 1.2089,
|
| 2307 |
+
"step": 326
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 0.5839285714285715,
|
| 2311 |
+
"grad_norm": 0.3254348933696747,
|
| 2312 |
+
"learning_rate": 7.623911680457198e-05,
|
| 2313 |
+
"loss": 1.0845,
|
| 2314 |
+
"step": 327
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 0.5857142857142857,
|
| 2318 |
+
"grad_norm": 0.3558744192123413,
|
| 2319 |
+
"learning_rate": 7.568466737947905e-05,
|
| 2320 |
+
"loss": 1.2339,
|
| 2321 |
+
"step": 328
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 0.5875,
|
| 2325 |
+
"grad_norm": 0.32738080620765686,
|
| 2326 |
+
"learning_rate": 7.513101128351454e-05,
|
| 2327 |
+
"loss": 0.8492,
|
| 2328 |
+
"step": 329
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 0.5892857142857143,
|
| 2332 |
+
"grad_norm": 0.36939212679862976,
|
| 2333 |
+
"learning_rate": 7.457816658065134e-05,
|
| 2334 |
+
"loss": 0.8973,
|
| 2335 |
+
"step": 330
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 0.5910714285714286,
|
| 2339 |
+
"grad_norm": 0.3393215835094452,
|
| 2340 |
+
"learning_rate": 7.402615130838917e-05,
|
| 2341 |
+
"loss": 1.0078,
|
| 2342 |
+
"step": 331
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 0.5928571428571429,
|
| 2346 |
+
"grad_norm": 0.402725487947464,
|
| 2347 |
+
"learning_rate": 7.347498347716624e-05,
|
| 2348 |
+
"loss": 1.2381,
|
| 2349 |
+
"step": 332
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 0.5946428571428571,
|
| 2353 |
+
"grad_norm": 0.33164989948272705,
|
| 2354 |
+
"learning_rate": 7.292468106977148e-05,
|
| 2355 |
+
"loss": 1.197,
|
| 2356 |
+
"step": 333
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 0.5964285714285714,
|
| 2360 |
+
"grad_norm": 0.3454689681529999,
|
| 2361 |
+
"learning_rate": 7.237526204075797e-05,
|
| 2362 |
+
"loss": 1.0244,
|
| 2363 |
+
"step": 334
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 0.5982142857142857,
|
| 2367 |
+
"grad_norm": 0.3584868907928467,
|
| 2368 |
+
"learning_rate": 7.182674431585704e-05,
|
| 2369 |
+
"loss": 1.062,
|
| 2370 |
+
"step": 335
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 0.6,
|
| 2374 |
+
"grad_norm": 0.3318020701408386,
|
| 2375 |
+
"learning_rate": 7.127914579139338e-05,
|
| 2376 |
+
"loss": 1.3696,
|
| 2377 |
+
"step": 336
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 0.6017857142857143,
|
| 2381 |
+
"grad_norm": 0.38772639632225037,
|
| 2382 |
+
"learning_rate": 7.073248433370124e-05,
|
| 2383 |
+
"loss": 1.1725,
|
| 2384 |
+
"step": 337
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 0.6035714285714285,
|
| 2388 |
+
"grad_norm": 0.33527833223342896,
|
| 2389 |
+
"learning_rate": 7.018677777854157e-05,
|
| 2390 |
+
"loss": 1.0849,
|
| 2391 |
+
"step": 338
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 0.6053571428571428,
|
| 2395 |
+
"grad_norm": 0.3958960473537445,
|
| 2396 |
+
"learning_rate": 6.964204393051981e-05,
|
| 2397 |
+
"loss": 0.8494,
|
| 2398 |
+
"step": 339
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 0.6071428571428571,
|
| 2402 |
+
"grad_norm": 0.3545495569705963,
|
| 2403 |
+
"learning_rate": 6.909830056250527e-05,
|
| 2404 |
+
"loss": 0.843,
|
| 2405 |
+
"step": 340
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 0.6089285714285714,
|
| 2409 |
+
"grad_norm": 0.3475857675075531,
|
| 2410 |
+
"learning_rate": 6.855556541505122e-05,
|
| 2411 |
+
"loss": 1.1228,
|
| 2412 |
+
"step": 341
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 0.6107142857142858,
|
| 2416 |
+
"grad_norm": 0.4085654616355896,
|
| 2417 |
+
"learning_rate": 6.801385619581592e-05,
|
| 2418 |
+
"loss": 0.8092,
|
| 2419 |
+
"step": 342
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 0.6125,
|
| 2423 |
+
"grad_norm": 0.4605577886104584,
|
| 2424 |
+
"learning_rate": 6.747319057898503e-05,
|
| 2425 |
+
"loss": 1.0999,
|
| 2426 |
+
"step": 343
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 0.6142857142857143,
|
| 2430 |
+
"grad_norm": 0.3840469717979431,
|
| 2431 |
+
"learning_rate": 6.693358620469487e-05,
|
| 2432 |
+
"loss": 1.2712,
|
| 2433 |
+
"step": 344
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 0.6160714285714286,
|
| 2437 |
+
"grad_norm": 0.3712100684642792,
|
| 2438 |
+
"learning_rate": 6.639506067845697e-05,
|
| 2439 |
+
"loss": 1.1401,
|
| 2440 |
+
"step": 345
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 0.6178571428571429,
|
| 2444 |
+
"grad_norm": 0.41670674085617065,
|
| 2445 |
+
"learning_rate": 6.585763157058358e-05,
|
| 2446 |
+
"loss": 1.151,
|
| 2447 |
+
"step": 346
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 0.6196428571428572,
|
| 2451 |
+
"grad_norm": 0.5912812352180481,
|
| 2452 |
+
"learning_rate": 6.53213164156144e-05,
|
| 2453 |
+
"loss": 1.447,
|
| 2454 |
+
"step": 347
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 0.6214285714285714,
|
| 2458 |
+
"grad_norm": 0.3995843231678009,
|
| 2459 |
+
"learning_rate": 6.478613271174453e-05,
|
| 2460 |
+
"loss": 1.6645,
|
| 2461 |
+
"step": 348
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 0.6232142857142857,
|
| 2465 |
+
"grad_norm": 0.5337942242622375,
|
| 2466 |
+
"learning_rate": 6.425209792025358e-05,
|
| 2467 |
+
"loss": 1.8703,
|
| 2468 |
+
"step": 349
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 0.625,
|
| 2472 |
+
"grad_norm": 0.5726771354675293,
|
| 2473 |
+
"learning_rate": 6.371922946493591e-05,
|
| 2474 |
+
"loss": 1.9795,
|
| 2475 |
+
"step": 350
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 0.6267857142857143,
|
| 2479 |
+
"grad_norm": 0.17994743585586548,
|
| 2480 |
+
"learning_rate": 6.318754473153221e-05,
|
| 2481 |
+
"loss": 1.5463,
|
| 2482 |
+
"step": 351
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 0.6285714285714286,
|
| 2486 |
+
"grad_norm": 0.16961722075939178,
|
| 2487 |
+
"learning_rate": 6.26570610671622e-05,
|
| 2488 |
+
"loss": 1.5388,
|
| 2489 |
+
"step": 352
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 0.6303571428571428,
|
| 2493 |
+
"grad_norm": 0.19410116970539093,
|
| 2494 |
+
"learning_rate": 6.21277957797587e-05,
|
| 2495 |
+
"loss": 1.7282,
|
| 2496 |
+
"step": 353
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 0.6321428571428571,
|
| 2500 |
+
"grad_norm": 0.1906638890504837,
|
| 2501 |
+
"learning_rate": 6.159976613750286e-05,
|
| 2502 |
+
"loss": 1.5167,
|
| 2503 |
+
"step": 354
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 0.6339285714285714,
|
| 2507 |
+
"grad_norm": 0.21918904781341553,
|
| 2508 |
+
"learning_rate": 6.107298936826086e-05,
|
| 2509 |
+
"loss": 1.7446,
|
| 2510 |
+
"step": 355
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 0.6357142857142857,
|
| 2514 |
+
"grad_norm": 0.19429104030132294,
|
| 2515 |
+
"learning_rate": 6.0547482659021706e-05,
|
| 2516 |
+
"loss": 1.7166,
|
| 2517 |
+
"step": 356
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 0.6375,
|
| 2521 |
+
"grad_norm": 0.21244099736213684,
|
| 2522 |
+
"learning_rate": 6.002326315533665e-05,
|
| 2523 |
+
"loss": 1.7319,
|
| 2524 |
+
"step": 357
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 0.6392857142857142,
|
| 2528 |
+
"grad_norm": 0.22793884575366974,
|
| 2529 |
+
"learning_rate": 5.950034796075947e-05,
|
| 2530 |
+
"loss": 1.6573,
|
| 2531 |
+
"step": 358
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 0.6410714285714286,
|
| 2535 |
+
"grad_norm": 0.23558048903942108,
|
| 2536 |
+
"learning_rate": 5.897875413628884e-05,
|
| 2537 |
+
"loss": 1.7359,
|
| 2538 |
+
"step": 359
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 0.6428571428571429,
|
| 2542 |
+
"grad_norm": 0.21951396763324738,
|
| 2543 |
+
"learning_rate": 5.845849869981137e-05,
|
| 2544 |
+
"loss": 1.6536,
|
| 2545 |
+
"step": 360
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 0.6446428571428572,
|
| 2549 |
+
"grad_norm": 0.22028475999832153,
|
| 2550 |
+
"learning_rate": 5.793959862554652e-05,
|
| 2551 |
+
"loss": 1.7257,
|
| 2552 |
+
"step": 361
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 0.6464285714285715,
|
| 2556 |
+
"grad_norm": 0.23070622980594635,
|
| 2557 |
+
"learning_rate": 5.7422070843492734e-05,
|
| 2558 |
+
"loss": 1.6149,
|
| 2559 |
+
"step": 362
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 0.6482142857142857,
|
| 2563 |
+
"grad_norm": 0.22408127784729004,
|
| 2564 |
+
"learning_rate": 5.6905932238875123e-05,
|
| 2565 |
+
"loss": 1.5936,
|
| 2566 |
+
"step": 363
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 0.65,
|
| 2570 |
+
"grad_norm": 0.23098969459533691,
|
| 2571 |
+
"learning_rate": 5.639119965159446e-05,
|
| 2572 |
+
"loss": 1.7313,
|
| 2573 |
+
"step": 364
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 0.6517857142857143,
|
| 2577 |
+
"grad_norm": 0.2533377707004547,
|
| 2578 |
+
"learning_rate": 5.5877889875677845e-05,
|
| 2579 |
+
"loss": 1.9422,
|
| 2580 |
+
"step": 365
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 0.6535714285714286,
|
| 2584 |
+
"grad_norm": 0.2515897750854492,
|
| 2585 |
+
"learning_rate": 5.5366019658730825e-05,
|
| 2586 |
+
"loss": 1.8331,
|
| 2587 |
+
"step": 366
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 0.6553571428571429,
|
| 2591 |
+
"grad_norm": 0.2757863700389862,
|
| 2592 |
+
"learning_rate": 5.485560570139061e-05,
|
| 2593 |
+
"loss": 1.7759,
|
| 2594 |
+
"step": 367
|
| 2595 |
+
},
|
| 2596 |
+
{
|
| 2597 |
+
"epoch": 0.6571428571428571,
|
| 2598 |
+
"grad_norm": 0.29774826765060425,
|
| 2599 |
+
"learning_rate": 5.434666465678175e-05,
|
| 2600 |
+
"loss": 1.334,
|
| 2601 |
+
"step": 368
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 0.6589285714285714,
|
| 2605 |
+
"grad_norm": 0.28664523363113403,
|
| 2606 |
+
"learning_rate": 5.383921312997242e-05,
|
| 2607 |
+
"loss": 1.6234,
|
| 2608 |
+
"step": 369
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 0.6607142857142857,
|
| 2612 |
+
"grad_norm": 0.30774933099746704,
|
| 2613 |
+
"learning_rate": 5.333326767743263e-05,
|
| 2614 |
+
"loss": 1.4743,
|
| 2615 |
+
"step": 370
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 0.6625,
|
| 2619 |
+
"grad_norm": 0.26094669103622437,
|
| 2620 |
+
"learning_rate": 5.282884480649435e-05,
|
| 2621 |
+
"loss": 1.1046,
|
| 2622 |
+
"step": 371
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 0.6642857142857143,
|
| 2626 |
+
"grad_norm": 0.2818247079849243,
|
| 2627 |
+
"learning_rate": 5.232596097481251e-05,
|
| 2628 |
+
"loss": 0.8417,
|
| 2629 |
+
"step": 372
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 0.6660714285714285,
|
| 2633 |
+
"grad_norm": 0.3089035749435425,
|
| 2634 |
+
"learning_rate": 5.182463258982846e-05,
|
| 2635 |
+
"loss": 1.3922,
|
| 2636 |
+
"step": 373
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 0.6678571428571428,
|
| 2640 |
+
"grad_norm": 0.2375502586364746,
|
| 2641 |
+
"learning_rate": 5.132487600823438e-05,
|
| 2642 |
+
"loss": 1.0855,
|
| 2643 |
+
"step": 374
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 0.6696428571428571,
|
| 2647 |
+
"grad_norm": 0.3417636454105377,
|
| 2648 |
+
"learning_rate": 5.082670753543961e-05,
|
| 2649 |
+
"loss": 1.0819,
|
| 2650 |
+
"step": 375
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 0.6714285714285714,
|
| 2654 |
+
"grad_norm": 0.2587840259075165,
|
| 2655 |
+
"learning_rate": 5.033014342503889e-05,
|
| 2656 |
+
"loss": 1.1154,
|
| 2657 |
+
"step": 376
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 0.6732142857142858,
|
| 2661 |
+
"grad_norm": 0.29829278588294983,
|
| 2662 |
+
"learning_rate": 4.9835199878281765e-05,
|
| 2663 |
+
"loss": 0.9634,
|
| 2664 |
+
"step": 377
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 0.675,
|
| 2668 |
+
"grad_norm": 0.307190477848053,
|
| 2669 |
+
"learning_rate": 4.9341893043544185e-05,
|
| 2670 |
+
"loss": 1.1533,
|
| 2671 |
+
"step": 378
|
| 2672 |
+
},
|
| 2673 |
+
{
|
| 2674 |
+
"epoch": 0.6767857142857143,
|
| 2675 |
+
"grad_norm": 0.3548847734928131,
|
| 2676 |
+
"learning_rate": 4.8850239015801625e-05,
|
| 2677 |
+
"loss": 1.2046,
|
| 2678 |
+
"step": 379
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 0.6785714285714286,
|
| 2682 |
+
"grad_norm": 0.3130282759666443,
|
| 2683 |
+
"learning_rate": 4.836025383610382e-05,
|
| 2684 |
+
"loss": 1.1391,
|
| 2685 |
+
"step": 380
|
| 2686 |
+
},
|
| 2687 |
+
{
|
| 2688 |
+
"epoch": 0.6803571428571429,
|
| 2689 |
+
"grad_norm": 0.3400501012802124,
|
| 2690 |
+
"learning_rate": 4.787195349105159e-05,
|
| 2691 |
+
"loss": 1.0226,
|
| 2692 |
+
"step": 381
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"epoch": 0.6821428571428572,
|
| 2696 |
+
"grad_norm": 0.3462565541267395,
|
| 2697 |
+
"learning_rate": 4.7385353912275165e-05,
|
| 2698 |
+
"loss": 1.0968,
|
| 2699 |
+
"step": 382
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 0.6839285714285714,
|
| 2703 |
+
"grad_norm": 0.331123948097229,
|
| 2704 |
+
"learning_rate": 4.690047097591427e-05,
|
| 2705 |
+
"loss": 1.1918,
|
| 2706 |
+
"step": 383
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 0.6857142857142857,
|
| 2710 |
+
"grad_norm": 0.354432612657547,
|
| 2711 |
+
"learning_rate": 4.6417320502100316e-05,
|
| 2712 |
+
"loss": 1.2261,
|
| 2713 |
+
"step": 384
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 0.6875,
|
| 2717 |
+
"grad_norm": 0.34844398498535156,
|
| 2718 |
+
"learning_rate": 4.593591825444028e-05,
|
| 2719 |
+
"loss": 0.7991,
|
| 2720 |
+
"step": 385
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 0.6892857142857143,
|
| 2724 |
+
"grad_norm": 0.3462367653846741,
|
| 2725 |
+
"learning_rate": 4.545627993950201e-05,
|
| 2726 |
+
"loss": 1.1343,
|
| 2727 |
+
"step": 386
|
| 2728 |
+
},
|
| 2729 |
+
{
|
| 2730 |
+
"epoch": 0.6910714285714286,
|
| 2731 |
+
"grad_norm": 0.3352709114551544,
|
| 2732 |
+
"learning_rate": 4.497842120630229e-05,
|
| 2733 |
+
"loss": 1.1023,
|
| 2734 |
+
"step": 387
|
| 2735 |
+
},
|
| 2736 |
+
{
|
| 2737 |
+
"epoch": 0.6928571428571428,
|
| 2738 |
+
"grad_norm": 0.3581717610359192,
|
| 2739 |
+
"learning_rate": 4.4502357645795976e-05,
|
| 2740 |
+
"loss": 0.9181,
|
| 2741 |
+
"step": 388
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 0.6946428571428571,
|
| 2745 |
+
"grad_norm": 0.35995498299598694,
|
| 2746 |
+
"learning_rate": 4.402810479036725e-05,
|
| 2747 |
+
"loss": 1.3445,
|
| 2748 |
+
"step": 389
|
| 2749 |
+
},
|
| 2750 |
+
{
|
| 2751 |
+
"epoch": 0.6964285714285714,
|
| 2752 |
+
"grad_norm": 0.372935950756073,
|
| 2753 |
+
"learning_rate": 4.355567811332311e-05,
|
| 2754 |
+
"loss": 1.0725,
|
| 2755 |
+
"step": 390
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 0.6982142857142857,
|
| 2759 |
+
"grad_norm": 0.3759123980998993,
|
| 2760 |
+
"learning_rate": 4.30850930283882e-05,
|
| 2761 |
+
"loss": 1.1824,
|
| 2762 |
+
"step": 391
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 0.7,
|
| 2766 |
+
"grad_norm": 0.391770601272583,
|
| 2767 |
+
"learning_rate": 4.2616364889202254e-05,
|
| 2768 |
+
"loss": 1.3516,
|
| 2769 |
+
"step": 392
|
| 2770 |
+
},
|
| 2771 |
+
{
|
| 2772 |
+
"epoch": 0.7017857142857142,
|
| 2773 |
+
"grad_norm": 0.3785625696182251,
|
| 2774 |
+
"learning_rate": 4.214950898881892e-05,
|
| 2775 |
+
"loss": 1.1624,
|
| 2776 |
+
"step": 393
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"epoch": 0.7035714285714286,
|
| 2780 |
+
"grad_norm": 0.4284125864505768,
|
| 2781 |
+
"learning_rate": 4.168454055920681e-05,
|
| 2782 |
+
"loss": 1.1318,
|
| 2783 |
+
"step": 394
|
| 2784 |
+
},
|
| 2785 |
+
{
|
| 2786 |
+
"epoch": 0.7053571428571429,
|
| 2787 |
+
"grad_norm": 0.391161173582077,
|
| 2788 |
+
"learning_rate": 4.12214747707527e-05,
|
| 2789 |
+
"loss": 1.4543,
|
| 2790 |
+
"step": 395
|
| 2791 |
+
},
|
| 2792 |
+
{
|
| 2793 |
+
"epoch": 0.7071428571428572,
|
| 2794 |
+
"grad_norm": 0.3723802864551544,
|
| 2795 |
+
"learning_rate": 4.0760326731766374e-05,
|
| 2796 |
+
"loss": 1.4265,
|
| 2797 |
+
"step": 396
|
| 2798 |
+
},
|
| 2799 |
+
{
|
| 2800 |
+
"epoch": 0.7089285714285715,
|
| 2801 |
+
"grad_norm": 0.4321235418319702,
|
| 2802 |
+
"learning_rate": 4.030111148798775e-05,
|
| 2803 |
+
"loss": 1.4523,
|
| 2804 |
+
"step": 397
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 0.7107142857142857,
|
| 2808 |
+
"grad_norm": 0.43963977694511414,
|
| 2809 |
+
"learning_rate": 3.9843844022096135e-05,
|
| 2810 |
+
"loss": 1.7426,
|
| 2811 |
+
"step": 398
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"epoch": 0.7125,
|
| 2815 |
+
"grad_norm": 0.5444914698600769,
|
| 2816 |
+
"learning_rate": 3.938853925322118e-05,
|
| 2817 |
+
"loss": 1.8907,
|
| 2818 |
+
"step": 399
|
| 2819 |
+
},
|
| 2820 |
+
{
|
| 2821 |
+
"epoch": 0.7142857142857143,
|
| 2822 |
+
"grad_norm": 0.9059175252914429,
|
| 2823 |
+
"learning_rate": 3.893521203645618e-05,
|
| 2824 |
+
"loss": 2.2147,
|
| 2825 |
+
"step": 400
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 0.7160714285714286,
|
| 2829 |
+
"grad_norm": 0.20250235497951508,
|
| 2830 |
+
"learning_rate": 3.848387716237353e-05,
|
| 2831 |
+
"loss": 1.7341,
|
| 2832 |
+
"step": 401
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 0.7178571428571429,
|
| 2836 |
+
"grad_norm": 0.1863853931427002,
|
| 2837 |
+
"learning_rate": 3.8034549356541894e-05,
|
| 2838 |
+
"loss": 1.6956,
|
| 2839 |
+
"step": 402
|
| 2840 |
+
},
|
| 2841 |
+
{
|
| 2842 |
+
"epoch": 0.7196428571428571,
|
| 2843 |
+
"grad_norm": 0.19317291676998138,
|
| 2844 |
+
"learning_rate": 3.7587243279046056e-05,
|
| 2845 |
+
"loss": 1.7165,
|
| 2846 |
+
"step": 403
|
| 2847 |
+
},
|
| 2848 |
+
{
|
| 2849 |
+
"epoch": 0.7214285714285714,
|
| 2850 |
+
"grad_norm": 0.21101966500282288,
|
| 2851 |
+
"learning_rate": 3.714197352400849e-05,
|
| 2852 |
+
"loss": 1.8306,
|
| 2853 |
+
"step": 404
|
| 2854 |
+
},
|
| 2855 |
+
{
|
| 2856 |
+
"epoch": 0.7232142857142857,
|
| 2857 |
+
"grad_norm": 0.22385361790657043,
|
| 2858 |
+
"learning_rate": 3.669875461911297e-05,
|
| 2859 |
+
"loss": 1.7104,
|
| 2860 |
+
"step": 405
|
| 2861 |
+
},
|
| 2862 |
+
{
|
| 2863 |
+
"epoch": 0.725,
|
| 2864 |
+
"grad_norm": 0.22555914521217346,
|
| 2865 |
+
"learning_rate": 3.6257601025131026e-05,
|
| 2866 |
+
"loss": 1.5668,
|
| 2867 |
+
"step": 406
|
| 2868 |
+
},
|
| 2869 |
+
{
|
| 2870 |
+
"epoch": 0.7267857142857143,
|
| 2871 |
+
"grad_norm": 0.21916812658309937,
|
| 2872 |
+
"learning_rate": 3.581852713544983e-05,
|
| 2873 |
+
"loss": 1.7827,
|
| 2874 |
+
"step": 407
|
| 2875 |
+
},
|
| 2876 |
+
{
|
| 2877 |
+
"epoch": 0.7285714285714285,
|
| 2878 |
+
"grad_norm": 0.23447498679161072,
|
| 2879 |
+
"learning_rate": 3.538154727560259e-05,
|
| 2880 |
+
"loss": 1.8308,
|
| 2881 |
+
"step": 408
|
| 2882 |
+
},
|
| 2883 |
+
{
|
| 2884 |
+
"epoch": 0.7303571428571428,
|
| 2885 |
+
"grad_norm": 0.21024593710899353,
|
| 2886 |
+
"learning_rate": 3.494667570280132e-05,
|
| 2887 |
+
"loss": 1.613,
|
| 2888 |
+
"step": 409
|
| 2889 |
+
},
|
| 2890 |
+
{
|
| 2891 |
+
"epoch": 0.7321428571428571,
|
| 2892 |
+
"grad_norm": 0.23882578313350677,
|
| 2893 |
+
"learning_rate": 3.45139266054715e-05,
|
| 2894 |
+
"loss": 1.6853,
|
| 2895 |
+
"step": 410
|
| 2896 |
+
},
|
| 2897 |
+
{
|
| 2898 |
+
"epoch": 0.7339285714285714,
|
| 2899 |
+
"grad_norm": 0.23162604868412018,
|
| 2900 |
+
"learning_rate": 3.408331410278929e-05,
|
| 2901 |
+
"loss": 1.7371,
|
| 2902 |
+
"step": 411
|
| 2903 |
+
},
|
| 2904 |
+
{
|
| 2905 |
+
"epoch": 0.7357142857142858,
|
| 2906 |
+
"grad_norm": 0.23150567710399628,
|
| 2907 |
+
"learning_rate": 3.3654852244220826e-05,
|
| 2908 |
+
"loss": 1.7505,
|
| 2909 |
+
"step": 412
|
| 2910 |
+
},
|
| 2911 |
+
{
|
| 2912 |
+
"epoch": 0.7375,
|
| 2913 |
+
"grad_norm": 0.23027552664279938,
|
| 2914 |
+
"learning_rate": 3.322855500906373e-05,
|
| 2915 |
+
"loss": 1.7128,
|
| 2916 |
+
"step": 413
|
| 2917 |
+
},
|
| 2918 |
+
{
|
| 2919 |
+
"epoch": 0.7392857142857143,
|
| 2920 |
+
"grad_norm": 0.22426114976406097,
|
| 2921 |
+
"learning_rate": 3.2804436305991214e-05,
|
| 2922 |
+
"loss": 1.7721,
|
| 2923 |
+
"step": 414
|
| 2924 |
+
},
|
| 2925 |
+
{
|
| 2926 |
+
"epoch": 0.7410714285714286,
|
| 2927 |
+
"grad_norm": 0.22792723774909973,
|
| 2928 |
+
"learning_rate": 3.238250997259808e-05,
|
| 2929 |
+
"loss": 1.7089,
|
| 2930 |
+
"step": 415
|
| 2931 |
+
},
|
| 2932 |
+
{
|
| 2933 |
+
"epoch": 0.7428571428571429,
|
| 2934 |
+
"grad_norm": 0.2450588494539261,
|
| 2935 |
+
"learning_rate": 3.196278977494934e-05,
|
| 2936 |
+
"loss": 1.744,
|
| 2937 |
+
"step": 416
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 0.7446428571428572,
|
| 2941 |
+
"grad_norm": 0.2348526120185852,
|
| 2942 |
+
"learning_rate": 3.154528940713113e-05,
|
| 2943 |
+
"loss": 1.8496,
|
| 2944 |
+
"step": 417
|
| 2945 |
+
},
|
| 2946 |
+
{
|
| 2947 |
+
"epoch": 0.7464285714285714,
|
| 2948 |
+
"grad_norm": 0.2519124746322632,
|
| 2949 |
+
"learning_rate": 3.113002249080386e-05,
|
| 2950 |
+
"loss": 1.76,
|
| 2951 |
+
"step": 418
|
| 2952 |
+
},
|
| 2953 |
+
{
|
| 2954 |
+
"epoch": 0.7482142857142857,
|
| 2955 |
+
"grad_norm": 0.27859431505203247,
|
| 2956 |
+
"learning_rate": 3.071700257475768e-05,
|
| 2957 |
+
"loss": 1.6493,
|
| 2958 |
+
"step": 419
|
| 2959 |
+
},
|
| 2960 |
+
{
|
| 2961 |
+
"epoch": 0.75,
|
| 2962 |
+
"grad_norm": 0.2539427876472473,
|
| 2963 |
+
"learning_rate": 3.030624313447067e-05,
|
| 2964 |
+
"loss": 1.7619,
|
| 2965 |
+
"step": 420
|
| 2966 |
+
},
|
| 2967 |
+
{
|
| 2968 |
+
"epoch": 0.75,
|
| 2969 |
+
"eval_loss": 1.4375773668289185,
|
| 2970 |
+
"eval_runtime": 13.3809,
|
| 2971 |
+
"eval_samples_per_second": 17.637,
|
| 2972 |
+
"eval_steps_per_second": 8.819,
|
| 2973 |
+
"step": 420
|
| 2974 |
}
|
| 2975 |
],
|
| 2976 |
"logging_steps": 1,
|
|
|
|
| 2990 |
"attributes": {}
|
| 2991 |
}
|
| 2992 |
},
|
| 2993 |
+
"total_flos": 1.3697781125559091e+17,
|
| 2994 |
"train_batch_size": 2,
|
| 2995 |
"trial_name": null,
|
| 2996 |
"trial_params": null
|