tuantmdev commited on
Commit
548de81
·
verified ·
1 Parent(s): 7834b93

Training in progress, step 250, 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:ec730630dd47ffcebec478577b0e29ac7ce5b94a93b5ddfadf8f8e2e05913a74
3
  size 335604696
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4f18440af0a47a3648e8e2493b41f8c69470684ab52a2f3f51fe86e66d9b485
3
  size 335604696
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c1be83eee156b64ccaa6715a3d74b963b801b2e16737814a4ba3c1b06bd3e964
3
  size 170920084
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e7f854548d6f6813f20a24c97ccc4bb7783db372e5705aa9ae1b608b31636d3
3
  size 170920084
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d5aab15a6bfd3da33b5a55e0d0c415be6212b4d8d069f7e06df29dcd13fa3a33
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eba5341e98f7a2a22d7405d74e4126c56a8d134d8178ce64467fc6df1c2be05
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:530505d607699f384741067a5f9139d72f043713adb680898a3f1b5714170c97
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5182c64d745614fbc9217a3122cffdaa4b9683ea7cfb2e6909f3ad8f175d0a03
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "best_metric": 1.6030395030975342,
3
- "best_model_checkpoint": "miner_id_24/checkpoint-200",
4
- "epoch": 0.21203286509408958,
5
  "eval_steps": 50,
6
- "global_step": 200,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -82,6 +82,21 @@
82
  "eval_samples_per_second": 14.5,
83
  "eval_steps_per_second": 7.259,
84
  "step": 200
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
85
  }
86
  ],
87
  "logging_steps": 40,
@@ -110,7 +125,7 @@
110
  "attributes": {}
111
  }
112
  },
113
- "total_flos": 7.156512927095194e+16,
114
  "train_batch_size": 2,
115
  "trial_name": null,
116
  "trial_params": null
 
1
  {
2
+ "best_metric": 1.5612239837646484,
3
+ "best_model_checkpoint": "miner_id_24/checkpoint-250",
4
+ "epoch": 0.26504108136761195,
5
  "eval_steps": 50,
6
+ "global_step": 250,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
82
  "eval_samples_per_second": 14.5,
83
  "eval_steps_per_second": 7.259,
84
  "step": 200
85
+ },
86
+ {
87
+ "epoch": 0.2544394381129075,
88
+ "grad_norm": 21.09881019592285,
89
+ "learning_rate": 5e-05,
90
+ "loss": 11.9366,
91
+ "step": 240
92
+ },
93
+ {
94
+ "epoch": 0.26504108136761195,
95
+ "eval_loss": 1.5612239837646484,
96
+ "eval_runtime": 54.8542,
97
+ "eval_samples_per_second": 14.493,
98
+ "eval_steps_per_second": 7.256,
99
+ "step": 250
100
  }
101
  ],
102
  "logging_steps": 40,
 
125
  "attributes": {}
126
  }
127
  },
128
+ "total_flos": 8.937905582820557e+16,
129
  "train_batch_size": 2,
130
  "trial_name": null,
131
  "trial_params": null