drzo commited on
Commit
51f495b
·
verified ·
1 Parent(s): 64134c7

Training update: run 598, 500 iters, cache_key=ci

Browse files
Files changed (3) hide show
  1. README.md +4 -4
  2. pytorch_model.bin +1 -1
  3. training_metadata.json +12 -12
README.md CHANGED
@@ -25,10 +25,10 @@ capabilities inspired by cognitive science and AGI research.
25
 
26
  ## Training Details
27
 
28
- - **Checkpoint ID**: ckpt_20260923_020847_1000_46032ab3_9470fbb7
29
- - **Training Iteration**: 1000
30
- - **Validation Loss**: 2.0753641605377195
31
- - **Quality Score**: 153600.42989232062
32
 
33
  ## Echo Self Features
34
 
 
25
 
26
  ## Training Details
27
 
28
+ - **Checkpoint ID**: ckpt_20260923_075250_1500_46032ab3_9470fbb7
29
+ - **Training Iteration**: 1500
30
+ - **Validation Loss**: 0.4853347837924957
31
+ - **Quality Score**: 230400.5738101433
32
 
33
  ## Echo Self Features
34
 
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:892f7c44dcd1c9f427228a462c2af7428a2fc10b174f2d0df18309f7fa96d1a0
3
  size 51464851
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71a0a1171765d5bca6ac3c473b207be2d1bfa83b577c7971e6dc6c6553135cfb
3
  size 51464851
training_metadata.json CHANGED
@@ -1,11 +1,11 @@
1
  {
2
- "checkpoint_id": "ckpt_20260923_020847_1000_46032ab3_9470fbb7",
3
- "created_at": "2026-09-23T02:08:47.350791",
4
- "iteration": 1000,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
- "val_loss": 2.0753641605377195,
8
- "learning_rate": 2.7972e-05,
9
  "model_config": {
10
  "n_layer": 4,
11
  "n_head": 4,
@@ -40,20 +40,20 @@
40
  "block_size": 1024
41
  },
42
  "metrics": {
43
- "val_loss": 2.0753641605377195,
44
- "connection_ratio": 0.5499999999999999,
45
- "tokens_processed": 2048000,
46
- "training_speed_iters_per_sec": 0.0716625040404187
47
  },
48
  "tags": [
49
  "phase_adaptive_mastery",
50
- "medium_quality",
51
  "nanecho",
52
  "curriculum",
53
  "introspection"
54
  ],
55
  "parent_checkpoint": null,
56
- "notes": "Training checkpoint at iteration 1000 (resumed from iteration 500) | Phase: adaptive_mastery",
57
  "file_size_mb": 276.6400327682495,
58
- "quality_score": 153600.42989232062
59
  }
 
1
  {
2
+ "checkpoint_id": "ckpt_20260923_075250_1500_46032ab3_9470fbb7",
3
+ "created_at": "2026-09-23T07:52:50.780667",
4
+ "iteration": 1500,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
+ "val_loss": 0.4853347837924957,
8
+ "learning_rate": 4.1972e-05,
9
  "model_config": {
10
  "n_layer": 4,
11
  "n_head": 4,
 
40
  "block_size": 1024
41
  },
42
  "metrics": {
43
+ "val_loss": 0.4853347837924957,
44
+ "connection_ratio": 0.8000000000000002,
45
+ "tokens_processed": 3072000,
46
+ "training_speed_iters_per_sec": 0.08181078579029949
47
  },
48
  "tags": [
49
  "phase_adaptive_mastery",
50
+ "high_quality",
51
  "nanecho",
52
  "curriculum",
53
  "introspection"
54
  ],
55
  "parent_checkpoint": null,
56
+ "notes": "Training checkpoint at iteration 1500 (resumed from iteration 1000) | Phase: adaptive_mastery",
57
  "file_size_mb": 276.6400327682495,
58
+ "quality_score": 230400.5738101433
59
  }