diff --git a/early_exit_20250817_layers_5_kl0_25/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2457b6c6de83d3ba8738bf82961275f19ba89133
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07837de4aac568f367b773a9fa7c83f63f00f79a6d55285ede23a8cd33f7e90b
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_1000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_1000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1c6a4ca40657ea68534d8178a676a5d43156b46a
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_1000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:010178105f916551ad0e5d249749cd4f1803a83364379b32ee016abcb12fb621
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_1500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_1500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..767d2bbbd920aa0da5676162d67e81d664d6b124
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_1500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83c6b229cea6a622adad4cd83d103ca4db4774600fa328fdb6e11c63d006f7b5
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_2000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_2000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b9334f0f277a7546bb5e7756d06d3611090467a4
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_2000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a1bad52e342572ba7f37faacbbb6db8850d2804f115672f53ebb2bacae55118
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_2500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_25/step_2500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b3cc00a04ed950e3113e419d288b894688ff67d3
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_2500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17ce5b5621baa8bc69de1301911eb87994c562c78e7a870e957f14154a691096
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_25/step_2500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_2500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a7bfd2bd96ca0caa2a83e192d16a4e0b97183029
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_2500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fca6ee0bfcb5f572d767cd95f743813ffa219b31e9ca7769cd597cdf90fb555
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_3000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_3000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..aefd46a26ebb4dee40baebd513995df033a91d11
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_3000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70ea8e8ac529c747fee3326a8bc3c0af36e9749fa7226346dbf2daf5d804abdc
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_3500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_3500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a01e47507dcb71dc0d9dbd06f32f2e2b1e0dc702
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_3500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:273af2867436e58e50844b183f496953a4322f2106174f65cdf8f4752710bfac
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_4000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_4000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..92bc9b829c8b666cf302dcd246662137e5c18d00
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_4000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6c504a4c9d3352c31e0583c783347cbea637d3ff9fcf44bcfacf13844cf2d5c
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_4500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_4500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c8cd6f1e4ae0a3b06062046513a9dce7c9ee2d83
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_4500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04963cfbcdc9cd2dc389e7e28a770239911ce9b334576d08b2893eb1694ee9b5
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_25/step_500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fca03244d525669c58b6ff8d26218be5b5ca372c
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f05a8849df30e7f4627109595f9ff10dede74411779cb8d35331b0bb76d19677
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_25/step_500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2b018ee8883c78dde983ae79043135a03746986b
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f360088c05bc92e8d3c4eabc91db63a3e2724e36c03e275f5cac29a042c09f0d
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_5000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_5000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b7d0d34b569d20b9ea835ca1c14cb1554816892f
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_5000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f428af5b13e6cc0afa86eb5ce37f81fe889be900ce84bbd46e0e4e322327284e
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_25/step_5500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_25/step_5500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f4e4e51a73058ebd3f12dc103ae37e9aab4e636d
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_5500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74449800989026130ff27bcae1274e725e5d5ca92814d200c2e706d9661a7107
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_25/step_5500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_25/step_5500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2398017c5e8a30f9cded81729bc7e74ad9fd0cdf
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_25/step_5500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45f16f652cbae6a07047c1a4051ae50fb9a5e60bc186065030005485e8d0caf8
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_1000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_1000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..476ef12bcbe791f75aa68fe3e69217c31d91a35c
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_1000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33738442e1247e815fb2c500345319a5563d04000652fb2e24ec3b99631a06d7
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_1000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_1000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2a0a7a5c42f191a85ba60738a5b43609aeae30b4
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_1000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d3fe9a2492170a213e876e838e3694e7e11842a1ce16a69a3ed4abd97c25142
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_1500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_1500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0a66a539b9e71e502c4d5c99fd64ba6078849789
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_1500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4fe2e83d79187c01397242bd31b08951ec9a17c2aa4835fbf217ae214dd6d30
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_1500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_1500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f85c490e436a1aa99e5355a87eeeee686d07495d
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_1500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4238b92fcab22412d76d75193cd25ef6ecce70cd3e73786b9a0a08e95b7827f4
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_2000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_2000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9e8da511b7475fbb84bbd85290a97560c4b0d166
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_2000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07e05a05fe400df853f1ef976108852b8cfe0295578167ed8d2a111866ccae22
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_2000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_2000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..91cbaa38c4a54af914bdc55679538639b48ae649
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_2000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fddb33783b5d1b9cf4a8746b02ff240ae871abf2dce3330173431ad59361248
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_2500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_2500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e0ba5b803677474ba27dbfe272ba563bc4f5ddba
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_2500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91c0aab502ac28109143132ed98849852ecd804a375003ebbd18fe1e52923b32
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_2500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_2500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..21a6266e152144fdc3a7f95ddb4e4035dfebedb6
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_2500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9971aa1129ada0584e6c8a881b75d64dc119a3651edb778fd0b1ec9e46ec7da
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_3000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_3000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6291852ace231623bc3c30ede59f864d31f4cd83
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_3000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a8bada79a15e7fc2e3273cdc44976607af137b163ba7c4d6b2ffa68f21c7953
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_3000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_3000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..bab4a716703faae3d62c8417d9be076453fed6d4
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_3000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:327823a4df4fc17121bca57111c7eae599c1f1a8da3557fe6062b04ef0f15218
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_3500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_3500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6625d5805aa70169cb18959ccda9aa36fbff3571
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_3500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab59da6ad5bfc19e5a30e46d734f6a96c6c39d0944edd864d2e2e9846d739865
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_3500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_3500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c7984987cea7e02018c9b2217b21388fc395e7c3
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_3500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05631d45d62b040edcaa295a2ac561aa67530432eaec2821d182d533d66ceeee
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_4000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_4000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..577d7d591ec215e33cd4b18046040677eea51e58
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_4000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90bce4068ab86d26f73141badfd7c856f5c9b1d1ccf8087fc4a1e6eff1043f8e
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_4000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_4000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dbcf80a64e753c8954820a039f5776f62ec89a53
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_4000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f6fca8b465a580576bb9f3ac7fa32011667086d054c3183f59159d1f3349b2e
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_4500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_4500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1c9f18f6cbc55c222e13b4a85813231330acbeef
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_4500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fadbb614b7b2e1ae2a48dbd34e074dbab0461f72fd0d28db48104551f001abb8
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_4500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_4500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8816e7aaf7ae0d168d1ea99005b760c1ead5d673
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_4500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:feac8841f6756b1981e6a780263c2680488263b65383c3b489cab58ee633cb9a
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl0_5/step_500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl0_5/step_500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..38e6d433b13d292716977a95ea768f6c9b565c75
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:233b532b92f23ae96259c7f005df971212582b60258c503bea00d10dee942ad5
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl0_5/step_500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl0_5/step_500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3cd62d747a5f82559cb70e2f828efd4a27bc5d96
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl0_5/step_500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cbf678583da0b9b3730d4d0cc24bd3d8fa2717f8bb7e850c073eee856ece348
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_1000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_1000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d4b59c6e313f764b9662a7496b8632d4fd63782c
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_1000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d35f90cc59f64cb79f5adca290507cbe08e0633d63d38c295c99e6062c78f2af
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_1000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_1000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ee00efc17dab61267ce7b792fb7c278b2c73bed4
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_1000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1546154ad4bd130bcfbcad7a9a31becf48dea2522da6ad21ee711c0e3ca8cacc
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_1500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_1500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1a3640b9a8e6373a9e87628b9f4d8deb7dfa1c08
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_1500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d8a81f3799794e66bb9b4bcea4f0d4e8b397eef8f497bc18e34fcb35a64477e
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_1500/early_exiter/adapter_config.json b/early_exit_20250817_layers_5_kl1_0/step_1500/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..3413e05cebb398a29a50b0bd9455aed509c56566
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_1500/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250817_layers_5_kl1_0/step_1500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_1500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..78e4d5624fdeca34ecab91d80b6d8638e3192a70
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_1500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c07cad6c04a6b209c5d30f349968e875250093ebacc2f5064fa5abe35d2c47f
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_2000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_2000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c4f7c3226c4cc2d1c8060f0605e11faae631242
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_2000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab939b589ea04dccc46699e69076fbcb4e563be64155220671927656e11ef621
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_2000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_2000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1adeed3c45b4ac47390a62c4663d0b039a33f2d5
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_2000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8aba5d4d6e46b2e580e3d233607894b8f0064fcb1d6cf3565a747de5f1abb8a
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_2500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_2500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6e091f15333ade706527e65ebdcd595e6404fc64
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_2500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3d0c199a29a0bd6e6e3d36923f986dc9e50cea94ca49c5f760b5aa9afa8ed47
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_2500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_2500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f5894328cdb4a7f9431d422e0666540220fdd18b
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_2500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90793ab06310160794c6cdaba44f64957b59ee903ca487c7e28ba9bfd3c2b970
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_3000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_3000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..640784c54b4925717a6d76d55c840d415760674c
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_3000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0bc46a32522eeb22fc65dda72ec2ea23a26d524ee6312685b2cee5999431f42
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_3000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_3000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..581e5bf80eade659c95eb48083db124713f7d453
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_3000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b75ffc50b1284f9b6ce5897a58dfa59448c51a4ca3b92fb3c68d50a7e4f9e2dc
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_3500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_3500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..175581cbf75b5635abbe745eb5c697d5de96459a
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_3500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e798c99ca4ef518c1ed0553c037dab3cccab3424a51d894cdb84d1b6c5d6773
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_3500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_3500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7ed050126466728364c26665d923549670a8bf5d
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_3500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3bad102aabb761b5ef950fb7864d5ffb697f128e45d37e97d331b1502e0de78
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_4000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_4000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..df3b16e7d749636b529482f906b979e24ef6d3e9
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_4000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aee0950af96d5b2e7519d64e713ac98af1b6df3714f39612874b686099f87c18
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_4000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_4000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7d71868f177a08f82aae48e4f44b4315f00bcbf2
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_4000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38eae4c6bbce7689260103bf71cd539d4b887c11dd61243026feb7972fc2b1ec
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0/step_500/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0/step_500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..350178442409912b52191eab424221e4ec4d6c24
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:463833d32818b0f56ee21cb0057493ee25c4500b2f66051075ca79a27894df29
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0/step_500/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0/step_500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..926560a88ece4258068b21ebf6ce7c18b49f59b3
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0/step_500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68fa34fabf333552510d2cb3a2fee13a55f30ac4f9d0b5eaeefb9c946c3cb8a6
+size 1884453408
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..65e6b61ebbaf0011e0caf3de6184e82938f3f533
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f75ffbaea7390f197d644bc3608b02731e7820c4121fab094c3cefc863775b7
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exiter/adapter_config.json b/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e9116cf5f47dbdbf5101cf322fa6ac870a51060d
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6e763405024b1e877661c3b3acabcae6a01c3ca1
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_1000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:feb8ba4607b1d0c0a675661b898e2fd5e48f19f7155c00580ce8776df94e5fd4
+size 1875736896
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_2000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0_rank8/step_2000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..86a1c542a0f2d39e416b81c231bfa5ccbb5232b9
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_2000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24f9fc755411bced0c877579f73136071e8ce8bd03670fba0708ac730e96ff13
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_2000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0_rank8/step_2000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9395dc13fb58907ccae96763adb57110a4709cda
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_2000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80cd4fed507cb81615ac9da7b696453c15b28ba3e888369ba09c82e92342c4e7
+size 1875736896
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_3000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0_rank8/step_3000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cc0db63de813c9db9803061032da67ecbd1960b4
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_3000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a93b6974addada5137c319602ee5e19c6fcb632c70ea3f1014bd553ef55a03
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_3000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0_rank8/step_3000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..19748f0ef53be725357d4e6fa493744ea81d98a7
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_3000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8da1ec24895e6ab0685adfb2daefa6f13d130b2d7e53243eab1a34ec995ea44c
+size 1875736896
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_4000/early_exit_probes.pt b/early_exit_20250817_layers_5_kl1_0_rank8/step_4000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a3bd40da0bcc7bc0bef629b3a96dc5145dffd3b5
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_4000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:feaacaccecf75cf2388a19129564f85d62baa723619722280736bf1dbc977546
+size 94502251
diff --git a/early_exit_20250817_layers_5_kl1_0_rank8/step_4000/early_exiter/adapter_model.safetensors b/early_exit_20250817_layers_5_kl1_0_rank8/step_4000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..236e7cb176f8f17171a4a4f3ff4eefa34030e7be
--- /dev/null
+++ b/early_exit_20250817_layers_5_kl1_0_rank8/step_4000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d00c0bc5282ceba103f972cca6057631a98b6c57388a837e6a989d3aafca5de3
+size 1875736896
diff --git a/early_exit_20250818_layers_5_kl1_0/README.md b/early_exit_20250818_layers_5_kl1_0/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..fb0893c2a4d79d9bc98c4f39e5b899f11c962dfb
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/README.md
@@ -0,0 +1,207 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.17.0
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0390ae791c0e859504cb57e5089062c206f0d9bc
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55c848725b820a88040d8ef47836383b83baa108da0c1d7b2b6177abc65c3498
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/early_exiter/adapter_config.json b/early_exit_20250818_layers_5_kl1_0/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..3413e05cebb398a29a50b0bd9455aed509c56566
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3a67de84416e5e710a0ccfb2d0511e61b2a052e1
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9d8f542a9a96dbf7983fa134ad535f7eee02d8162be4b8b450e4d576250519a
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_100/README.md b/early_exit_20250818_layers_5_kl1_0/step_100/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..fb0893c2a4d79d9bc98c4f39e5b899f11c962dfb
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_100/README.md
@@ -0,0 +1,207 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.17.0
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_100/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_100/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..16b802634ab7719a0fa2b62c49793d38d435054c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_100/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac52f57797a315cc1515a72a4498ad0d4ce1a48394d3c2fd0f485755f2f84acd
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_100/early_exiter/adapter_config.json b/early_exit_20250818_layers_5_kl1_0/step_100/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..b14a1f397b567057c5b3a21b52a67eb75ac69d8c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_100/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_100/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_100/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fc43f0b33e77c165a5040328f2fa5bfe90bfa5b3
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_100/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f180b0118e8c2385b62813be5a92dca2a6aebff8c6dbd2010d7c77a6a66ab508
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_100/metadata.json b/early_exit_20250818_layers_5_kl1_0/step_100/metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..4453fca5915cf996beb71cc25dfa4e16aa14609c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_100/metadata.json
@@ -0,0 +1,13 @@
+{
+  "base_model_name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "exitable_layer_idxs": [
+    5.0,
+    10.0,
+    15.0,
+    20.0,
+    25.0,
+    Infinity
+  ],
+  "total_exitable_layers": 5,
+  "has_early_exit_probes": true
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_2000/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_2000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..206dbe58e22e459300df83583737ac4587ad0ca0
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_2000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a75853c38d136f7ae8e88f641a14828d3fed6723d633f8030a17e7fd70ea8d2
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_2000/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_2000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..317bde6ef8f893c35043d6d3a63010deb60bee48
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_2000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00b9303ba75f33d7f282e1bcf9e7ae9916047aa49b9061211ec7f0551548ab78
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_2500/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_2500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d6a8de4bd4dbce3bf2ee12d9173df280ab1d81e1
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_2500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0399d565a5f0548174a6a806b4a4ef426b33cd61558f7f4f623379a073839ffb
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_2500/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_2500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1da829a95cf47ed57d728fcaa4dc38a3edde677d
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_2500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6db7ed96a849eb3ffa848e9a1ecbc9dccef938a54fdf1155178308f3423989e0
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_3000/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_3000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b74263149630f9dda242f9b9c4d4f65fd7880204
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_3000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6481afcf339f72aece3854daa682f37686530f46e0b8b6a812780a8e1e167856
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_3500/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_3500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bc8bf0a050dba8b0c2504a38d8c041a945229c02
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_3500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:886f70693f31f3e37ca12dc3486844e4b2756a2494bdbb743da80420feafb821
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_3500/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_3500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b4b73310f145ab5e806187e6187a64030933ee8b
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_3500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7301bbe4c5d7ad9f5ee5f9f55dc666023d389c2ab9e55667277235d0bce7b46
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_4000/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_4000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..91b089007c25ee74768717ebba5f9bff7b92e67a
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_4000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726a23350376409b896f1704feca0b171cad9f5895cf291cf77c23f6459b2826
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_4000/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_4000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4cb0f924ebb1cc34311b7d535277dc9029bf4c85
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_4000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b571779ae4ac51248bed62bb332cd8d924e07db07533c69360c867774b4594f5
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_4500/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_4500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f535a03a5aa2b0e8f859a6a64a80e40a2821dfc3
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_4500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00e7301175872fe838f85773bba225356ab479f3e485bd83450840b4b20ae87a
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_4500/early_exiter/adapter_config.json b/early_exit_20250818_layers_5_kl1_0/step_4500/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..3413e05cebb398a29a50b0bd9455aed509c56566
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_4500/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_4500/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_4500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0fd495b2fbb34b4f5c1304664927488b57a3ee68
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_4500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9108b69dad0c1683f7b68bdfc91a8f65496b4958ce9f18431fdb4497e1e4085a
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_50/README.md b/early_exit_20250818_layers_5_kl1_0/step_50/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..fb0893c2a4d79d9bc98c4f39e5b899f11c962dfb
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_50/README.md
@@ -0,0 +1,207 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.17.0
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_50/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_50/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c59ff3c80db7a4ec66086acc70aaedeb2fa851a4
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_50/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a84cc01e7821644b7c279f2b78ba076ede0f8df05803ea3dddd8c8df9641f89
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_50/early_exiter/adapter_config.json b/early_exit_20250818_layers_5_kl1_0/step_50/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..b14a1f397b567057c5b3a21b52a67eb75ac69d8c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_50/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_50/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_50/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..598f5e65eb1da5fc6cd0af38cbd52bb08debe95b
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_50/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a60b692b1258ae8e7ccaa556c713c22b638726a3722f8d2a3d809ff3b1425065
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_50/metadata.json b/early_exit_20250818_layers_5_kl1_0/step_50/metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..4453fca5915cf996beb71cc25dfa4e16aa14609c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_50/metadata.json
@@ -0,0 +1,13 @@
+{
+  "base_model_name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "exitable_layer_idxs": [
+    5.0,
+    10.0,
+    15.0,
+    20.0,
+    25.0,
+    Infinity
+  ],
+  "total_exitable_layers": 5,
+  "has_early_exit_probes": true
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5000/README.md b/early_exit_20250818_layers_5_kl1_0/step_5000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..fb0893c2a4d79d9bc98c4f39e5b899f11c962dfb
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5000/README.md
@@ -0,0 +1,207 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.17.0
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5000/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_5000/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..612add248162e86063d2e7b920ad27a8813f7680
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5000/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b724c1ace2c8f6df3b93f06f111b8042e60797132b26714b295cc3bfed0474f
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5000/early_exiter/adapter_config.json b/early_exit_20250818_layers_5_kl1_0/step_5000/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..3413e05cebb398a29a50b0bd9455aed509c56566
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5000/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5000/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_5000/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0b65a46f80c945b5f16c4509be6fe73e213d427e
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5000/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6b25ed2444ded1e2928182a8dc37eb64407ddd1c357624dbe6cae866d518f9d
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5000/metadata.json b/early_exit_20250818_layers_5_kl1_0/step_5000/metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..4453fca5915cf996beb71cc25dfa4e16aa14609c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5000/metadata.json
@@ -0,0 +1,13 @@
+{
+  "base_model_name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "exitable_layer_idxs": [
+    5.0,
+    10.0,
+    15.0,
+    20.0,
+    25.0,
+    Infinity
+  ],
+  "total_exitable_layers": 5,
+  "has_early_exit_probes": true
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5500/README.md b/early_exit_20250818_layers_5_kl1_0/step_5500/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..fb0893c2a4d79d9bc98c4f39e5b899f11c962dfb
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5500/README.md
@@ -0,0 +1,207 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.17.0
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5500/early_exit_probes.pt b/early_exit_20250818_layers_5_kl1_0/step_5500/early_exit_probes.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0302d093d4e18857c5066537919017bef2a233e1
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5500/early_exit_probes.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1efe2e370137c36b1682f9d980f7a9fb51dc9ab0fb8c4901ecf8393a059df236
+size 94502251
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5500/early_exiter/adapter_config.json b/early_exit_20250818_layers_5_kl1_0/step_5500/early_exiter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..3413e05cebb398a29a50b0bd9455aed509c56566
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5500/early_exiter/adapter_config.json
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5500/early_exiter/adapter_model.safetensors b/early_exit_20250818_layers_5_kl1_0/step_5500/early_exiter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1158d7b89cbb05e713b5dacc2db3fcdaba2f3c57
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5500/early_exiter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8da9f9bc4015e80389197e8371902edbb35969bfbba3da100bca36c202ffe01c
+size 1884453408
diff --git a/early_exit_20250818_layers_5_kl1_0/step_5500/metadata.json b/early_exit_20250818_layers_5_kl1_0/step_5500/metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..4453fca5915cf996beb71cc25dfa4e16aa14609c
--- /dev/null
+++ b/early_exit_20250818_layers_5_kl1_0/step_5500/metadata.json
@@ -0,0 +1,13 @@
+{
+  "base_model_name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+  "exitable_layer_idxs": [
+    5.0,
+    10.0,
+    15.0,
+    20.0,
+    25.0,
+    Infinity
+  ],
+  "total_exitable_layers": 5,
+  "has_early_exit_probes": true
+}
\ No newline at end of file