filipesantoscv11 commited on
Commit
ec399a4
·
verified ·
1 Parent(s): 15be5ad

Training in progress, step 30, checkpoint

Browse files
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6e177764fc2976bc3a17add480660888b63d9bd435a9217553a4cde8142690b6
3
  size 10636352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:836189b9e784d4f857624d445dd0d18799d241b6bc7ebd21ff6a676ff6d4bf54
3
  size 10636352
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fffc7954a2c0c6d42f97c5ca25ad0451196849d5f9a4c98814f574a8c0342426
3
  size 21354682
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77c70eccde7ac98b2656ed1df06da7bd47bbeb523173b78e25f5c0b30b0c6632
3
  size 21354682
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c953f40073a0e88b46dfd2c697cf1a9fd8434bb76cfee81236c986cc3bfff88e
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:006b30ad57b816573d29a4422a3a4a0960a035e484c7234a9432f1808f5e310e
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46d68217ddf3e899bf1495ee33b9ec14e2c91ed912ffcb460a7bb01192b04b7d
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0e9cc72c20ddd925ef39b6005e82a4d8730b1dde32cfcd070d74c83a8a3564a
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": 0.017185821697099892,
5
  "eval_steps": 5,
6
- "global_step": 20,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -89,6 +89,50 @@
89
  "eval_samples_per_second": 137.677,
90
  "eval_steps_per_second": 68.839,
91
  "step": 20
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
92
  }
93
  ],
94
  "logging_steps": 3,
@@ -103,12 +147,12 @@
103
  "should_evaluate": false,
104
  "should_log": false,
105
  "should_save": true,
106
- "should_training_stop": false
107
  },
108
  "attributes": {}
109
  }
110
  },
111
- "total_flos": 92689341087744.0,
112
  "train_batch_size": 2,
113
  "trial_name": null,
114
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.02577873254564984,
5
  "eval_steps": 5,
6
+ "global_step": 30,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
89
  "eval_samples_per_second": 137.677,
90
  "eval_steps_per_second": 68.839,
91
  "step": 20
92
+ },
93
+ {
94
+ "epoch": 0.01804511278195489,
95
+ "grad_norm": 3.3421783447265625,
96
+ "learning_rate": 4.2178276747988446e-05,
97
+ "loss": 8.2177,
98
+ "step": 21
99
+ },
100
+ {
101
+ "epoch": 0.02062298603651987,
102
+ "grad_norm": 3.7299230098724365,
103
+ "learning_rate": 2.061073738537635e-05,
104
+ "loss": 8.6419,
105
+ "step": 24
106
+ },
107
+ {
108
+ "epoch": 0.021482277121374866,
109
+ "eval_loss": 2.0797598361968994,
110
+ "eval_runtime": 3.71,
111
+ "eval_samples_per_second": 132.077,
112
+ "eval_steps_per_second": 66.038,
113
+ "step": 25
114
+ },
115
+ {
116
+ "epoch": 0.023200859291084855,
117
+ "grad_norm": 3.310062885284424,
118
+ "learning_rate": 5.449673790581611e-06,
119
+ "loss": 8.6889,
120
+ "step": 27
121
+ },
122
+ {
123
+ "epoch": 0.02577873254564984,
124
+ "grad_norm": 3.9792568683624268,
125
+ "learning_rate": 0.0,
126
+ "loss": 8.4295,
127
+ "step": 30
128
+ },
129
+ {
130
+ "epoch": 0.02577873254564984,
131
+ "eval_loss": 2.0769786834716797,
132
+ "eval_runtime": 3.4884,
133
+ "eval_samples_per_second": 140.464,
134
+ "eval_steps_per_second": 70.232,
135
+ "step": 30
136
  }
137
  ],
138
  "logging_steps": 3,
 
147
  "should_evaluate": false,
148
  "should_log": false,
149
  "should_save": true,
150
+ "should_training_stop": true
151
  },
152
  "attributes": {}
153
  }
154
  },
155
+ "total_flos": 137956228595712.0,
156
  "train_batch_size": 2,
157
  "trial_name": null,
158
  "trial_params": null