Hugging Face's logo

besimray
/
test

PEFT
Safetensors
llama
axolotl
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files
xet
 Community
besimray commited on
Commit
d8340e1
·
verified ·
1 Parent(s): f5ea875

Training in progress, step 10, checkpoint

Browse files
Files changed (5) hide show
  1. last-checkpoint/adapter_model.safetensors +1 -1
  2. last-checkpoint/optimizer.pt +1 -1
  3. last-checkpoint/rng_state.pth +1 -1
  4. last-checkpoint/scheduler.pt +1 -1
  5. last-checkpoint/trainer_state.json +46 -3
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:43dfae7d10f75e0524be41b70b288f5192eb29f95438d6e89ee1c45e33b79903
3
  size 45118424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:581ce02141e78d9221ba813009b4df857f792b7625b643e6f087a4d344bb7e7d
3
  size 45118424
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b26c7be4c47f5a3142696ddc782972c339d74a9dd867a80385a1a311981c07eb
3
  size 23159290
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a80cf6546c77389f1922dc367e78237cd9c56c5e0270814378fe278c0898905b
3
  size 23159290
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:332cde8a34859807112eada791b022852a0dfccfbfa4ed9dc03857068b9813a9
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6488621cfae2ffe0d34573d69d34b876de946ff462260a38dfb343677f810145
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cda2d7da3ce8a95be6df505b84eace6ccd5aa18ffc6d1bcc9a79572045c8d78f
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb578e75c11a81e85dda67a691f96ba4793a02960f1409fd3e1511aac873491a
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": 0.10526315789473684,
5
  "eval_steps": 8,
6
- "global_step": 5,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -50,6 +50,49 @@
50
  "learning_rate": 0.0001,
51
  "loss": 1.24,
52
  "step": 5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
53
  }
54
  ],
55
  "logging_steps": 1,
@@ -69,7 +112,7 @@
69
  "attributes": {}
70
  }
71
  },
72
- "total_flos": 498981949931520.0,
73
  "train_batch_size": 10,
74
  "trial_name": null,
75
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.21052631578947367,
5
  "eval_steps": 8,
6
+ "global_step": 10,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
50
  "learning_rate": 0.0001,
51
  "loss": 1.24,
52
  "step": 5
53
+ },
54
+ {
55
+ "epoch": 0.12631578947368421,
56
+ "grad_norm": 0.514263927936554,
57
+ "learning_rate": 0.00012,
58
+ "loss": 1.3899,
59
+ "step": 6
60
+ },
61
+ {
62
+ "epoch": 0.14736842105263157,
63
+ "grad_norm": 0.6221964955329895,
64
+ "learning_rate": 0.00014,
65
+ "loss": 1.3389,
66
+ "step": 7
67
+ },
68
+ {
69
+ "epoch": 0.16842105263157894,
70
+ "grad_norm": 0.4577077031135559,
71
+ "learning_rate": 0.00016,
72
+ "loss": 1.223,
73
+ "step": 8
74
+ },
75
+ {
76
+ "epoch": 0.16842105263157894,
77
+ "eval_loss": 1.2106529474258423,
78
+ "eval_runtime": 2.1078,
79
+ "eval_samples_per_second": 47.443,
80
+ "eval_steps_per_second": 4.744,
81
+ "step": 8
82
+ },
83
+ {
84
+ "epoch": 0.18947368421052632,
85
+ "grad_norm": 0.5334281921386719,
86
+ "learning_rate": 0.00018,
87
+ "loss": 1.2047,
88
+ "step": 9
89
+ },
90
+ {
91
+ "epoch": 0.21052631578947367,
92
+ "grad_norm": 0.44930824637413025,
93
+ "learning_rate": 0.0002,
94
+ "loss": 1.3736,
95
+ "step": 10
96
  }
97
  ],
98
  "logging_steps": 1,
 
112
  "attributes": {}
113
  }
114
  },
115
+ "total_flos": 975282902138880.0,
116
  "train_batch_size": 10,
117
  "trial_name": null,
118
  "trial_params": null