vovaRL commited on
Commit
bfe502a
·
verified ·
1 Parent(s): b76afcf

step 20000 | vovaRL | 5242880000 tok

Browse files
Files changed (3) hide show
  1. model.safetensors +1 -1
  2. optimizer.pt +1 -1
  3. training_state.json +48 -48
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:45809c30767929ab5704187e93adb326c84ea81f7f374b3c67eeeedc04027524
3
  size 5167420536
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49e6f7557d2eeae7c3437e524335b6113290dfbabfc4a7d2ff27b91d6633538b
3
  size 5167420536
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4967f30ccccc652a526376b8f384572ebcdbacd1702334db4c6208184b458551
3
  size 10333957008
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bd6d3835d3e27da9ce246f993f3bb35f09ff873453725efef511b437ebfbdb9
3
  size 10333957008
training_state.json CHANGED
@@ -1,70 +1,70 @@
1
  {
2
- "step": 18000,
3
- "global_tokens": 4718592000,
4
  "contributor": "vovaRL",
5
  "cloud": "home",
6
  "base_model": "scratch:colibri-nano",
7
- "wall_clock_s": 24694.61459493637,
8
  "history_tail": [
9
  {
10
- "step": 17960,
11
- "t": 1785217619.521,
12
- "ntp": 2.2682926654815674,
13
- "zloss": 0.2846938669681549,
14
- "lr": 3.846693141859465e-05,
15
- "grad_norm": 0.287109375,
16
- "ppl": 9.662888938603704,
17
- "balance_max_imbalance": 0.078369140625,
18
  "balance_dead_frac": 0.0,
19
- "tokens": 4708106240
20
  },
21
  {
22
- "step": 17970,
23
- "t": 1785217742.574,
24
- "ntp": 2.6745243072509766,
25
- "zloss": 0.2764131724834442,
26
- "lr": 3.8384993827393804e-05,
27
- "grad_norm": 0.291015625,
28
- "ppl": 14.505448061693928,
29
- "balance_max_imbalance": 0.072265625,
30
  "balance_dead_frac": 0.0,
31
- "tokens": 4710727680
32
  },
33
  {
34
- "step": 17980,
35
- "t": 1785217865.518,
36
- "ntp": 2.026712656021118,
37
- "zloss": 0.3065721392631531,
38
- "lr": 3.830344192745995e-05,
39
- "grad_norm": 0.28515625,
40
- "ppl": 7.589097334097592,
41
- "balance_max_imbalance": 0.06500244140625,
42
  "balance_dead_frac": 0.0,
43
- "tokens": 4713349120
44
  },
45
  {
46
- "step": 17990,
47
- "t": 1785217988.4,
48
- "ntp": 2.2962148189544678,
49
- "zloss": 0.27432844042778015,
50
- "lr": 3.822227596721427e-05,
51
- "grad_norm": 0.3125,
52
- "ppl": 9.936499731399175,
53
- "balance_max_imbalance": 0.1422119140625,
54
  "balance_dead_frac": 0.0,
55
- "tokens": 4715970560
56
  },
57
  {
58
- "step": 18000,
59
- "t": 1785218111.404,
60
- "ntp": 2.513723373413086,
61
- "zloss": 0.2994578182697296,
62
- "lr": 3.814149619390237e-05,
63
- "grad_norm": 0.265625,
64
- "ppl": 12.35083131066547,
65
- "balance_max_imbalance": 0.0731201171875,
66
  "balance_dead_frac": 0.0,
67
- "tokens": 4718592000
68
  }
69
  ]
70
  }
 
1
  {
2
+ "step": 20000,
3
+ "global_tokens": 5242880000,
4
  "contributor": "vovaRL",
5
  "cloud": "home",
6
  "base_model": "scratch:colibri-nano",
7
+ "wall_clock_s": 24702.277401924133,
8
  "history_tail": [
9
  {
10
+ "step": 19960,
11
+ "t": 1785242935.469,
12
+ "ntp": 2.2264950275421143,
13
+ "zloss": 0.28825539350509644,
14
+ "lr": 3.000328985477171e-05,
15
+ "grad_norm": 0.28515625,
16
+ "ppl": 9.267327361978355,
17
+ "balance_max_imbalance": 0.1207275390625,
18
  "balance_dead_frac": 0.0,
19
+ "tokens": 5232394240
20
  },
21
  {
22
+ "step": 19970,
23
+ "t": 1785243058.314,
24
+ "ntp": 2.558842420578003,
25
+ "zloss": 0.3002420961856842,
26
+ "lr": 3.000185054659739e-05,
27
+ "grad_norm": 0.302734375,
28
+ "ppl": 12.920851743202324,
29
+ "balance_max_imbalance": 0.06561279296875,
30
  "balance_dead_frac": 0.0,
31
+ "tokens": 5235015680
32
  },
33
  {
34
+ "step": 19980,
35
+ "t": 1785243181.299,
36
+ "ntp": 2.413142442703247,
37
+ "zloss": 0.2963441014289856,
38
+ "lr": 3.0000822466198297e-05,
39
+ "grad_norm": 0.2734375,
40
+ "ppl": 11.169004011286283,
41
+ "balance_max_imbalance": 0.09515380859375,
42
  "balance_dead_frac": 0.0,
43
+ "tokens": 5237637120
44
  },
45
  {
46
+ "step": 19990,
47
+ "t": 1785243304.446,
48
+ "ntp": 2.3685061931610107,
49
+ "zloss": 0.292042076587677,
50
+ "lr": 3.000020561670616e-05,
51
+ "grad_norm": 0.291015625,
52
+ "ppl": 10.681424375812318,
53
+ "balance_max_imbalance": 0.2261962890625,
54
  "balance_dead_frac": 0.0,
55
+ "tokens": 5240258560
56
  },
57
  {
58
+ "step": 20000,
59
+ "t": 1785243427.437,
60
+ "ntp": 2.834693670272827,
61
+ "zloss": 0.2698419988155365,
62
+ "lr": 2.9999999999999997e-05,
63
+ "grad_norm": 0.28125,
64
+ "ppl": 17.025184181485457,
65
+ "balance_max_imbalance": 0.1982421875,
66
  "balance_dead_frac": 0.0,
67
+ "tokens": 5242880000
68
  }
69
  ]
70
  }