drzo commited on
Commit
0253541
·
verified ·
1 Parent(s): c0d2a1e

Training update: run 482, 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 +10 -10
README.md CHANGED
@@ -25,10 +25,10 @@ capabilities inspired by cognitive science and AGI research.
25
 
26
  ## Training Details
27
 
28
- - **Checkpoint ID**: ckpt_20260525_021002_2000_22deff1b_9470fbb7
29
- - **Training Iteration**: 2000
30
- - **Validation Loss**: 0.8281021404266358
31
- - **Quality Score**: 307200.52635321725
32
 
33
  ## Echo Self Features
34
 
 
25
 
26
  ## Training Details
27
 
28
+ - **Checkpoint ID**: ckpt_20260525_084612_2500_22deff1b_9470fbb7
29
+ - **Training Iteration**: 2500
30
+ - **Validation Loss**: 0.13270145431160926
31
+ - **Quality Score**: 384000.71117398003
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 sha256:51ef633d3322889144d71f84c8217b33ee17586f4dbaa8e7d1347f4ac8a5a903
3
  size 51464851
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cad3c819296ffeb1a2ef04c96c84c84fa69fb0304f975f9330f278292b874a58
3
  size 51464851
training_metadata.json CHANGED
@@ -1,11 +1,11 @@
1
  {
2
- "checkpoint_id": "ckpt_20260525_021002_2000_22deff1b_9470fbb7",
3
- "created_at": "2026-05-25T02:10:02.386546",
4
- "iteration": 2000,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
- "val_loss": 0.8281021404266358,
8
- "learning_rate": 5.5972000000000006e-05,
9
  "model_config": {
10
  "n_layer": 4,
11
  "n_head": 4,
@@ -36,10 +36,10 @@
36
  "block_size": 1024
37
  },
38
  "metrics": {
39
- "val_loss": 0.8281021404266358,
40
  "connection_ratio": 1.0,
41
- "tokens_processed": 4096000,
42
- "training_speed_iters_per_sec": 0.08446381067758409
43
  },
44
  "tags": [
45
  "phase_adaptive_mastery",
@@ -49,7 +49,7 @@
49
  "introspection"
50
  ],
51
  "parent_checkpoint": null,
52
- "notes": "Training checkpoint at iteration 2000 (resumed from iteration 1500) | Phase: adaptive_mastery",
53
  "file_size_mb": 249.29106044769287,
54
- "quality_score": 307200.52635321725
55
  }
 
1
  {
2
+ "checkpoint_id": "ckpt_20260525_084612_2500_22deff1b_9470fbb7",
3
+ "created_at": "2026-05-25T08:46:12.396129",
4
+ "iteration": 2500,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
+ "val_loss": 0.13270145431160926,
8
+ "learning_rate": 6.9972e-05,
9
  "model_config": {
10
  "n_layer": 4,
11
  "n_head": 4,
 
36
  "block_size": 1024
37
  },
38
  "metrics": {
39
+ "val_loss": 0.13270145431160926,
40
  "connection_ratio": 1.0,
41
+ "tokens_processed": 5120000,
42
+ "training_speed_iters_per_sec": 0.1097300918481657
43
  },
44
  "tags": [
45
  "phase_adaptive_mastery",
 
49
  "introspection"
50
  ],
51
  "parent_checkpoint": null,
52
+ "notes": "Training checkpoint at iteration 2500 (resumed from iteration 2000) | Phase: adaptive_mastery",
53
  "file_size_mb": 249.29106044769287,
54
+ "quality_score": 384000.71117398003
55
  }