Second checkpoint
Browse files- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +27 -3
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 609389712
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ddf4492601f1de7b3bbcd918e30af5b791b82a6881c4651142c1a89beb90880a
|
3 |
size 609389712
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 43127132
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ddcff1d5bdbd8789dba1706a78bd01ddb67631093c37b89688236f340f238a00
|
3 |
size 43127132
|
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 sha256:46a442c3ca436aa6a10e4093e0aaf8d54298771a4565ad23c34571d316886c86
|
3 |
size 14244
|
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 sha256:29ff92e07eb84e01189de138508ab139b01aad8541db4bc42c48d76d71ae0b56
|
3 |
size 1064
|
trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -79,6 +79,30 @@
|
|
79 |
"learning_rate": 2.186973905723906e-05,
|
80 |
"loss": 3.6035,
|
81 |
"step": 600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
82 |
}
|
83 |
],
|
84 |
"logging_steps": 50,
|
@@ -86,7 +110,7 @@
|
|
86 |
"num_input_tokens_seen": 0,
|
87 |
"num_train_epochs": 1,
|
88 |
"save_steps": 100,
|
89 |
-
"total_flos": 1.
|
90 |
"train_batch_size": 8,
|
91 |
"trial_name": null,
|
92 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.16815554387808723,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 800,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
79 |
"learning_rate": 2.186973905723906e-05,
|
80 |
"loss": 3.6035,
|
81 |
"step": 600
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.14,
|
85 |
+
"learning_rate": 2.160669191919192e-05,
|
86 |
+
"loss": 3.5998,
|
87 |
+
"step": 650
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.15,
|
91 |
+
"learning_rate": 2.1343644781144782e-05,
|
92 |
+
"loss": 3.5809,
|
93 |
+
"step": 700
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.16,
|
97 |
+
"learning_rate": 2.1080597643097644e-05,
|
98 |
+
"loss": 3.5888,
|
99 |
+
"step": 750
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.17,
|
103 |
+
"learning_rate": 2.0817550505050505e-05,
|
104 |
+
"loss": 3.6076,
|
105 |
+
"step": 800
|
106 |
}
|
107 |
],
|
108 |
"logging_steps": 50,
|
|
|
110 |
"num_input_tokens_seen": 0,
|
111 |
"num_train_epochs": 1,
|
112 |
"save_steps": 100,
|
113 |
+
"total_flos": 1.9644023431710966e+18,
|
114 |
"train_batch_size": 8,
|
115 |
"trial_name": null,
|
116 |
"trial_params": null
|