drzo commited on
Commit
eb81cfc
·
verified ·
1 Parent(s): 9d7fcda

Training update: run 440, 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_20260514_151950_2000_22deff1b_9470fbb7
29
- - **Training Iteration**: 2000
30
- - **Validation Loss**: 0.7789712142944336
31
- - **Quality Score**: 307200.53274635813
32
 
33
  ## Echo Self Features
34
 
 
25
 
26
  ## Training Details
27
 
28
+ - **Checkpoint ID**: ckpt_20260514_205329_2500_22deff1b_9470fbb7
29
+ - **Training Iteration**: 2500
30
+ - **Validation Loss**: 0.15082413524389268
31
+ - **Quality Score**: 384000.7024516982
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:f6a8f3bdd7150dc7e2068f14551c2cfecbdbc76bf2e6ee490f18d1e64a66266e
3
  size 51464851
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da246f0c5c97c757b1bd6008315983b8cb4bc7bb709ac9a4a5d757f6d11fe3d9
3
  size 51464851
training_metadata.json CHANGED
@@ -1,11 +1,11 @@
1
  {
2
- "checkpoint_id": "ckpt_20260514_151950_2000_22deff1b_9470fbb7",
3
- "created_at": "2026-05-14T15:19:50.382830",
4
- "iteration": 2000,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
- "val_loss": 0.7789712142944336,
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": 0.7789712142944336,
40
  "connection_ratio": 1.0,
41
- "tokens_processed": 4096000,
42
- "training_speed_iters_per_sec": 0.07783556369346298
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": 307200.53274635813
55
  }
 
1
  {
2
+ "checkpoint_id": "ckpt_20260514_205329_2500_22deff1b_9470fbb7",
3
+ "created_at": "2026-05-14T20:53:29.614576",
4
+ "iteration": 2500,
5
  "epoch": 0,
6
  "train_loss": 0.0,
7
+ "val_loss": 0.15082413524389268,
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": 0.15082413524389268,
40
  "connection_ratio": 1.0,
41
+ "tokens_processed": 5120000,
42
+ "training_speed_iters_per_sec": 0.10506851655622926
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": 384000.7024516982
55
  }