Upload folder using huggingface_hub
Browse files- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +17 -3
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 184584072
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1fbe9be840c3e73d01b1f9b9d8bad30388f39399877a1a0baefcf3626870ef85
|
3 |
size 184584072
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 94020308
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f848ac8e584150ab43b1268aa56968de3223fa82881374212e2e73eeebbe84fb
|
3 |
size 94020308
|
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:a5f9c8466246784adca37141843ae311a771e871cf72714bcbcade1fbddb7478
|
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:e0d7c5182b2f032288a44ca4175101f97fe1e7458f1f586d7bf701dac390454d
|
3 |
size 1064
|
trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 1.
|
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,
|
@@ -147,6 +147,20 @@
|
|
147 |
"learning_rate": 8.225609429353187e-06,
|
148 |
"loss": 2.026,
|
149 |
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
150 |
}
|
151 |
],
|
152 |
"logging_steps": 5,
|
@@ -166,7 +180,7 @@
|
|
166 |
"attributes": {}
|
167 |
}
|
168 |
},
|
169 |
-
"total_flos": 2.
|
170 |
"train_batch_size": 1,
|
171 |
"trial_name": null,
|
172 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.8049281314168377,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 110,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
147 |
"learning_rate": 8.225609429353187e-06,
|
148 |
"loss": 2.026,
|
149 |
"step": 100
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.7227926078028748,
|
153 |
+
"grad_norm": 1.6016583442687988,
|
154 |
+
"learning_rate": 4.684610648167503e-06,
|
155 |
+
"loss": 2.0659,
|
156 |
+
"step": 105
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 1.8049281314168377,
|
160 |
+
"grad_norm": 1.5633082389831543,
|
161 |
+
"learning_rate": 2.100524384225555e-06,
|
162 |
+
"loss": 1.9925,
|
163 |
+
"step": 110
|
164 |
}
|
165 |
],
|
166 |
"logging_steps": 5,
|
|
|
180 |
"attributes": {}
|
181 |
}
|
182 |
},
|
183 |
+
"total_flos": 2.785615979702477e+16,
|
184 |
"train_batch_size": 1,
|
185 |
"trial_name": null,
|
186 |
"trial_params": null
|