Synth-Ecosystem commited on
Commit
03d5cea
·
verified ·
1 Parent(s): b153076

Training in progress, step 100, 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:e33944d49867f2fc809cfe620f2482ae5e9598d2b88f08b5894d6b655e50e08e
3
  size 38572808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c87e7474950b1ee16ecd75548c93aea36a344980521b979a0b97c51914d02942
3
  size 38572808
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:00aa55a299442d2890eb4f9068e2d158e53ff38a6913f260ae6b8f8d4bab7275
3
  size 39471301
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6585f555a0c09669158c8ffe6632792273eec8e2438412a5045d637c13549374
3
  size 39471301
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c580658143e6f42e7bf435905aa120c8fcbea49d122ca48ac719b751c7dbdc74
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0ce48f7838ef9c6db5df2d123e0c99865ee5b89de2a5ed04fe910b70ca2b12c
3
  size 14645
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e1fa56069dfc97761886a74a15b1ef7f748bf16c15bdb5df861362f49c2f527e
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a973502662ad7c5b97bc0ff30c444f8b3459d8504353d705bc19cbe98e9e4af3
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.08,
6
  "eval_steps": 500,
7
- "global_step": 50,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -58,6 +58,56 @@
58
  "mean_token_accuracy": 0.5868253096938133,
59
  "num_tokens": 120513.0,
60
  "step": 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
  }
62
  ],
63
  "logging_steps": 10,
@@ -77,7 +127,7 @@
77
  "attributes": {}
78
  }
79
  },
80
- "total_flos": 1344338096947200.0,
81
  "train_batch_size": 2,
82
  "trial_name": null,
83
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.16,
6
  "eval_steps": 500,
7
+ "global_step": 100,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
58
  "mean_token_accuracy": 0.5868253096938133,
59
  "num_tokens": 120513.0,
60
  "step": 50
61
+ },
62
+ {
63
+ "entropy": 2.52132785320282,
64
+ "epoch": 0.096,
65
+ "grad_norm": 2.265625,
66
+ "learning_rate": 0.00014460000000000002,
67
+ "loss": 2.5625015258789063,
68
+ "mean_token_accuracy": 0.5602172784507274,
69
+ "num_tokens": 145386.0,
70
+ "step": 60
71
+ },
72
+ {
73
+ "entropy": 2.4292581140995027,
74
+ "epoch": 0.112,
75
+ "grad_norm": 2.671875,
76
+ "learning_rate": 0.0001386,
77
+ "loss": 2.5273454666137694,
78
+ "mean_token_accuracy": 0.5630464181303978,
79
+ "num_tokens": 171111.0,
80
+ "step": 70
81
+ },
82
+ {
83
+ "entropy": 2.053872545063496,
84
+ "epoch": 0.128,
85
+ "grad_norm": 2.28125,
86
+ "learning_rate": 0.00013260000000000002,
87
+ "loss": 2.1423622131347657,
88
+ "mean_token_accuracy": 0.6182360425591469,
89
+ "num_tokens": 195172.0,
90
+ "step": 80
91
+ },
92
+ {
93
+ "entropy": 1.8883190169930457,
94
+ "epoch": 0.144,
95
+ "grad_norm": 2.046875,
96
+ "learning_rate": 0.00012660000000000001,
97
+ "loss": 1.9964679718017577,
98
+ "mean_token_accuracy": 0.635661780834198,
99
+ "num_tokens": 218330.0,
100
+ "step": 90
101
+ },
102
+ {
103
+ "entropy": 1.9393853038549422,
104
+ "epoch": 0.16,
105
+ "grad_norm": 2.40625,
106
+ "learning_rate": 0.00012060000000000002,
107
+ "loss": 2.022273063659668,
108
+ "mean_token_accuracy": 0.6229535043239594,
109
+ "num_tokens": 244458.0,
110
+ "step": 100
111
  }
112
  ],
113
  "logging_steps": 10,
 
127
  "attributes": {}
128
  }
129
  },
130
+ "total_flos": 2714607316131840.0,
131
  "train_batch_size": 2,
132
  "trial_name": null,
133
  "trial_params": null