mudassiruddin21 commited on
Commit
0347f1b
·
verified ·
1 Parent(s): 4fc7ef0

Run r2: logs and results

Browse files
runs/r2/logs/extract.log CHANGED
@@ -1 +1,19 @@
1
  [17:03:05] training files to featurize: ['chat.jsonl', 'code.jsonl', 'longform.jsonl', 'math.jsonl'] (0 earlier shards kept)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  [17:03:05] training files to featurize: ['chat.jsonl', 'code.jsonl', 'longform.jsonl', 'math.jsonl'] (0 earlier shards kept)
2
+ [transformers] `causal_conv1d_fn` is falling back to its reference PyTorch implementation because `causal_conv1d` is not installed. This is correct but much slower; install `causal_conv1d` for the optimized kernel.
3
+ [transformers] `chunk_gated_delta_rule` is falling back to its reference PyTorch implementation because `flash-linear-attention` is not installed. This is correct but much slower; install `flash-linear-attention` for the optimized kernel.
4
+ [17:03:20] target loaded in 15.0s (hidden states via 'base')
5
+ [17:03:51] dev features: 160 sequences in 32s; teacher-forced agreement with the engine's greedy outputs {'code': 0.9987, 'chat': 0.9943, 'math': 0.9976, 'longform': 0.9903}
6
+ [17:03:58] simulated baseline acceptance 2.792 vs kit-measured 2.802; per domain chat 3.065/3.075, code 3.749/3.773, math 3.625/3.656, longform 2.355/2.364; step accuracies [0.843, 0.77, 0.765]
7
+ [17:03:58] 9634 generated sequences to featurize
8
+ [17:04:41] group 0: 106336 generated tokens featurized, 2503 tok/s
9
+ [17:05:26] group 1: 220363 generated tokens featurized, 2519 tok/s
10
+ [17:06:09] group 2: 330106 generated tokens featurized, 2527 tok/s
11
+ [17:06:51] group 3: 434153 generated tokens featurized, 2517 tok/s
12
+ [17:07:36] group 4: 549771 generated tokens featurized, 2523 tok/s
13
+ [17:08:20] group 5: 657694 generated tokens featurized, 2519 tok/s
14
+ [17:09:00] group 6: 755023 generated tokens featurized, 2505 tok/s
15
+ [17:09:43] group 7: 863841 generated tokens featurized, 2505 tok/s
16
+ [17:10:25] group 8: 967864 generated tokens featurized, 2504 tok/s
17
+ [17:11:07] group 9: 1072461 generated tokens featurized, 2500 tok/s
18
+ [17:11:51] group 10: 1181336 generated tokens featurized, 2500 tok/s
19
+ [17:12:37] group 11: 1293591 generated tokens featurized, 2494 tok/s
runs/r2/logs/stats.log CHANGED
@@ -20,3 +20,4 @@
20
  [2026-11-18 17:02:59] ram 0.3/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.3/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 83 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
21
  [2026-11-18 17:03:02] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 62 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
22
  [2026-11-18 17:03:08] ram 0.6/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.6/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 61 W | load 0.0 | largest python 1.0 GB, python 0.1 GB, python 0.1 GB
 
 
20
  [2026-11-18 17:02:59] ram 0.3/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.3/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 83 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
21
  [2026-11-18 17:03:02] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 62 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
22
  [2026-11-18 17:03:08] ram 0.6/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.6/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 61 W | load 0.0 | largest python 1.0 GB, python 0.1 GB, python 0.1 GB
23
+ [2026-11-18 17:13:10] ram 2.0/172 GB used (cache 0.1, shmem 0.0) | cgroup 2.0/172 GB | disk 82 GB used | gpu 87.7/95.6 GiB, 100%, 443 W | load 0.0 | largest python 3.0 GB, python 0.1 GB, python 0.1 GB
runs/r2/logs/sync.log CHANGED
@@ -11,3 +11,4 @@ uploaded logs/results to mudassiruddin21/EMC:runs/r2
11
  uploaded logs/results to mudassiruddin21/EMC:runs/r2
12
  uploaded logs/results to mudassiruddin21/EMC:runs/r2
13
  uploaded logs/results to mudassiruddin21/EMC:runs/r2
 
 
11
  uploaded logs/results to mudassiruddin21/EMC:runs/r2
12
  uploaded logs/results to mudassiruddin21/EMC:runs/r2
13
  uploaded logs/results to mudassiruddin21/EMC:runs/r2
14
+ uploaded logs/results to mudassiruddin21/EMC:runs/r2
runs/r2/results/extract_checks.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "kept_train_shards": 0,
3
+ "files": [
4
+ "chat.jsonl",
5
+ "code.jsonl",
6
+ "longform.jsonl",
7
+ "math.jsonl"
8
+ ],
9
+ "load_seconds": 15.0,
10
+ "hidden_state_call": "base",
11
+ "dev_teacher_forced_agreement": {
12
+ "code": 0.9987,
13
+ "chat": 0.9943,
14
+ "math": 0.9976,
15
+ "longform": 0.9903
16
+ },
17
+ "sim_baseline_dev": {
18
+ "acceptance_by_domain": {
19
+ "code": 3.748700173310225,
20
+ "chat": 3.065347721822542,
21
+ "math": 3.6251830161054173,
22
+ "longform": 2.355075648188838
23
+ },
24
+ "acceptance_length": 2.792494400176257,
25
+ "step_accuracy": [
26
+ 0.8426951494159017,
27
+ 0.7702044333046134,
28
+ 0.7650262079092819
29
+ ],
30
+ "rounds": 27233,
31
+ "sequences": 160
32
+ },
33
+ "kit_baseline_dev": {
34
+ "acceptance_length": 2.8023814806597143,
35
+ "acceptance_by_domain": {
36
+ "chat": 3.0752094829348047,
37
+ "code": 3.7726477789436865,
38
+ "math": 3.655878859857482,
39
+ "longform": 2.363773395302098
40
+ }
41
+ }
42
+ }