Stage 1 complete: 10K steps, CKA layer alignment final checkpoint
Browse files- README.md +3 -3
- stage1_checkpoint.pt +2 -2
- stage1_metadata.json +5 -5
README.md
CHANGED
|
@@ -58,9 +58,9 @@ Key research findings applied:
|
|
| 58 |
## Current Checkpoint
|
| 59 |
|
| 60 |
- **Stage**: 1 (layer-alignment)
|
| 61 |
-
- **Step**:
|
| 62 |
-
- **Loss**: 0.
|
| 63 |
-
- **Updated**: 2026-03-12T22:
|
| 64 |
|
| 65 |
## Languages
|
| 66 |
|
|
|
|
| 58 |
## Current Checkpoint
|
| 59 |
|
| 60 |
- **Stage**: 1 (layer-alignment)
|
| 61 |
+
- **Step**: 10000
|
| 62 |
+
- **Loss**: 0.0109
|
| 63 |
+
- **Updated**: 2026-03-12T22:44:33.839474+00:00
|
| 64 |
|
| 65 |
## Languages
|
| 66 |
|
stage1_checkpoint.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b0b800a8a1f4121416a6b3695262449fff317eaa7be8401c1c5637a49ee6d40
|
| 3 |
+
size 1644727015
|
stage1_metadata.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"step":
|
| 3 |
-
"checkpoint_file": "
|
| 4 |
-
"timestamp": "2026-03-12T22:
|
| 5 |
"cka_history_keys": [
|
| 6 |
"0_0",
|
| 7 |
"1_1",
|
|
@@ -9,8 +9,8 @@
|
|
| 9 |
"4_3",
|
| 10 |
"6_4"
|
| 11 |
],
|
| 12 |
-
"final_loss": 0.
|
| 13 |
-
"best_loss": 0.
|
| 14 |
"training_config": {
|
| 15 |
"loss_type": "mse+cosine",
|
| 16 |
"cka_threshold": 0.75,
|
|
|
|
| 1 |
{
|
| 2 |
+
"step": 10000,
|
| 3 |
+
"checkpoint_file": "final.pt",
|
| 4 |
+
"timestamp": "2026-03-12T22:44:33.839474+00:00",
|
| 5 |
"cka_history_keys": [
|
| 6 |
"0_0",
|
| 7 |
"1_1",
|
|
|
|
| 9 |
"4_3",
|
| 10 |
"6_4"
|
| 11 |
],
|
| 12 |
+
"final_loss": 0.010861422750167549,
|
| 13 |
+
"best_loss": 0.010707496083341539,
|
| 14 |
"training_config": {
|
| 15 |
"loss_type": "mse+cosine",
|
| 16 |
"cka_threshold": 0.75,
|