wayyresearch
/

aetheris

Text Generation

state-space-model

mixture-of-experts

Mixture of Experts

knowledge-distillation

hybrid-architecture

Model card Files Files and versions

rcgalbo commited on Mar 12

Commit

ccac9cc

·

verified ·

1 Parent(s): ba9c5e2

Stage 1 checkpoint: [Step 50/20000] loss=7.7500

Files changed (3) hide show

README.md +3 -3
stage1_checkpoint.pt +2 -2
stage1_metadata.json +5 -5

README.md CHANGED Viewed

@@ -58,9 +58,9 @@ Key research findings applied:
 ## Current Checkpoint
 - **Stage**: 1 (layer-alignment)
-- **Step**: 10000
-- **Loss**: 0.0109
-- **Updated**: 2026-03-12T22:44:33.839474+00:00
 ## Languages

 ## Current Checkpoint
 - **Stage**: 1 (layer-alignment)
+- **Step**: 9000
+- **Loss**: 0.0126
+- **Updated**: 2026-03-12T22:48:49.369182+00:00
 ## Languages

stage1_checkpoint.pt CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b0b800a8a1f4121416a6b3695262449fff317eaa7be8401c1c5637a49ee6d40
-size 1644727015

 version https://git-lfs.github.com/spec/v1
+oid sha256:df3f45979da4c94b3a77c26e9bad87b03af7423887c55c2784853c9e96adeb6a
+size 1644745483

stage1_metadata.json CHANGED Viewed

@@ -1,7 +1,7 @@
 {
-  "step": 10000,
-  "checkpoint_file": "final.pt",
-  "timestamp": "2026-03-12T22:44:33.839474+00:00",
   "cka_history_keys": [
     "0_0",
     "1_1",
@@ -9,8 +9,8 @@
     "4_3",
     "6_4"
   ],
-  "final_loss": 0.010861422750167549,
-  "best_loss": 0.010707496083341539,
   "training_config": {
     "loss_type": "mse+cosine",
     "cka_threshold": 0.75,

 {
+  "step": 9000,
+  "checkpoint_file": "step_9000.pt",
+  "timestamp": "2026-03-12T22:48:49.369182+00:00",
   "cka_history_keys": [
     "0_0",
     "1_1",
     "4_3",
     "6_4"
   ],
+  "final_loss": 0.012602645438164473,
+  "best_loss": 0.011982748401351273,
   "training_config": {
     "loss_type": "mse+cosine",
     "cka_threshold": 0.75,