drzo commited on
Commit
75539f1
·
verified ·
1 Parent(s): dde1da7

Training update: run 600, 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_20260923_132749_2000_46032ab3_9470fbb7
29
- - **Training Iteration**: 2000
30
- - **Validation Loss**: 0.1407655218243599
31
- - **Quality Score**: 307200.7072636593
32
 
33
  ## Echo Self Features
34
 
 
25
 
26
  ## Training Details
27
 
28
+ - **Checkpoint ID**: ckpt_20260923_193341_2500_46032ab3_9470fbb7
29
+ - **Training Iteration**: 2500
30
+ - **Validation Loss**: 0.0046207179967314
31
+ - **Quality Score**: 384000.7794502122
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:aca00b45a088c188db8f0d12d304c245bff0f0d5dc9c2d90a29862ecd92655f3
3
  size 51464851
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:998d2edb3d6d4eb436c6ab43dde3c762e1ac2048977e26d8edfd4d19328af284
3
  size 51464851
training_metadata.json CHANGED
@@ -1,11 +1,11 @@
1
  {
2
- "checkpoint_id": "ckpt_20260923_132749_2000_46032ab3_9470fbb7",
3
- "created_at": "2026-09-23T13:27:49.556628",
4
- "iteration": 2000,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
- "val_loss": 0.1407655218243599,
8
- "learning_rate": 5.5972000000000006e-05,
9
  "model_config": {
10
  "n_layer": 4,
11
  "n_head": 4,
@@ -40,10 +40,10 @@
40
  "block_size": 1024
41
  },
42
  "metrics": {
43
- "val_loss": 0.1407655218243599,
44
  "connection_ratio": 1.0,
45
- "tokens_processed": 4096000,
46
- "training_speed_iters_per_sec": 0.10777615547142712
47
  },
48
  "tags": [
49
  "phase_adaptive_mastery",
@@ -53,7 +53,7 @@
53
  "introspection"
54
  ],
55
  "parent_checkpoint": null,
56
- "notes": "Training checkpoint at iteration 2000 (resumed from iteration 1500) | Phase: adaptive_mastery",
57
  "file_size_mb": 276.6400327682495,
58
- "quality_score": 307200.7072636593
59
  }
 
1
  {
2
+ "checkpoint_id": "ckpt_20260923_193341_2500_46032ab3_9470fbb7",
3
+ "created_at": "2026-09-23T19:33:41.341766",
4
+ "iteration": 2500,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
+ "val_loss": 0.0046207179967314,
8
+ "learning_rate": 6.9972e-05,
9
  "model_config": {
10
  "n_layer": 4,
11
  "n_head": 4,
 
40
  "block_size": 1024
41
  },
42
  "metrics": {
43
+ "val_loss": 0.0046207179967314,
44
  "connection_ratio": 1.0,
45
+ "tokens_processed": 5120000,
46
+ "training_speed_iters_per_sec": 0.0976437864922841
47
  },
48
  "tags": [
49
  "phase_adaptive_mastery",
 
53
  "introspection"
54
  ],
55
  "parent_checkpoint": null,
56
+ "notes": "Training checkpoint at iteration 2500 (resumed from iteration 2000) | Phase: adaptive_mastery",
57
  "file_size_mb": 276.6400327682495,
58
+ "quality_score": 384000.7794502122
59
  }