JacksonBrune commited on
Commit
6132528
·
verified ·
1 Parent(s): e7e910f

Training in progress, step 26, checkpoint

Browse files
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:024a060f41d65601143c24870507e53f6ddab95aa648175029904e922c30cc3e
3
  size 66103768
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44db42b174ddaf1ad51121894491110db5bb32b1c92d70469d4f722b9e63b025
3
  size 66103768
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fe6b7fe9a83ce1a8ecb369725994dab7e6aeeb03d0084f091e9d8a6e867a99f9
3
  size 33939412
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72a41f532ae0fa530776151a4255eec1ef96469ac4ced9cb587b7121087ede87
3
  size 33939412
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ae0af8e5d5a2bcd829f39a3e49fe731050968700411fa08d173724db21223734
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7ad3bc3eb66a1dafc7cac8d068083d1b3673a0556f86a48cc5f333f029b48f7
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fec74e92563bbe4b510ec939dc1c78d591826119d26043cc49e0a868b67121a2
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59af5f5d66ecb18e9abb84f5d5074a625502d490e59b29de1e9974286a967cbf
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": 0.005528973950026581,
5
  "eval_steps": 13,
6
- "global_step": 13,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -30,6 +30,21 @@
30
  "eval_samples_per_second": 19.002,
31
  "eval_steps_per_second": 9.501,
32
  "step": 13
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  }
34
  ],
35
  "logging_steps": 10,
@@ -49,7 +64,7 @@
49
  "attributes": {}
50
  }
51
  },
52
- "total_flos": 2952800325709824.0,
53
  "train_batch_size": 2,
54
  "trial_name": null,
55
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.011057947900053162,
5
  "eval_steps": 13,
6
+ "global_step": 26,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
30
  "eval_samples_per_second": 19.002,
31
  "eval_steps_per_second": 9.501,
32
  "step": 13
33
+ },
34
+ {
35
+ "epoch": 0.008506113769271665,
36
+ "grad_norm": 2.187023639678955,
37
+ "learning_rate": 0.00015000000000000001,
38
+ "loss": 6.2868,
39
+ "step": 20
40
+ },
41
+ {
42
+ "epoch": 0.011057947900053162,
43
+ "eval_loss": 1.4111206531524658,
44
+ "eval_runtime": 52.0517,
45
+ "eval_samples_per_second": 19.02,
46
+ "eval_steps_per_second": 9.51,
47
+ "step": 26
48
  }
49
  ],
50
  "logging_steps": 10,
 
64
  "attributes": {}
65
  }
66
  },
67
+ "total_flos": 5977620171558912.0,
68
  "train_batch_size": 2,
69
  "trial_name": null,
70
  "trial_params": null