Vishwas1
/

hummingbird_Llama_distill_float32

Vishwas1 commited on Feb 12

Commit

491e7bc

verified ·

1 Parent(s): d7496e8

Training finished (float32, Xavier Init), epochs=3, lr=1e-07, clip=1, batch_size=4, loss=8.6898, perplexity=nan

Files changed (1) hide show

student_model.pth CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f24a6805811ef34105013a1ccc234f2efff14104bccd8c00e9cc6574260862c3
-size 90951887

 version https://git-lfs.github.com/spec/v1
+oid sha256:d22463c5cf58a2336256656d65cb6e4f30adab7efa29f3fcdff3566deb6c1b30
+size 90954581