Upload depth-scaled transformer probes

layer_12.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f3f5efe333ba714404f30034cbedc8f8560e55a63ff19568bdab6a949f533f6
+size 6137752

layer_12_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 12,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 20,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 2.4374186965942384,
+  "eval_mse": 2.0876001670920714
+}

layer_16.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d63f9a3185a7d08d3f8adaaec5d57c269131507484916e5576adf6f0a1dc21e2
+size 5333000

layer_16_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 16,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 16,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 3.0125887276649475,
+  "eval_mse": 3.479990794368405
+}

layer_20.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b43049f2dc20f16b83b44c664b4edc59177dbb4a89cf3d6756cb3f6267832a16
+size 4528256

layer_20_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 20,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 12,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 3.406930178833008,
+  "eval_mse": 3.4354089314601683
+}

layer_24.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cfe4724fc2c87c898a756d65a9fd02f343edb00899f6a666a69f9d15ad3a047
+size 3723528

layer_24_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 24,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 8,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 4.520931562042236,
+  "eval_mse": 4.617096263317011
+}

layer_28.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62ab317791dd40865432f4453906b9ee3b1f1897b0a1c6ff32c1f5d6dbb283be
+size 2918832

layer_28_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 28,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 4,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 8.369592458343506,
+  "eval_mse": 8.405994867279654
+}

layer_32.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa8735474f1673449fbac2a61b4dcf8f8b328eca6cd302f50b533369cec60de1
+size 2315304

layer_32_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 32,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 1,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 0.45685000762939454,
+  "eval_mse": 0.4507866852244911
+}

layer_8.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1fb5d91b8ae2473a9aa374f65336c92aaebfc8a752ecb9b56b6dcf8044c18ad
+size 6942496

layer_8_config.json

@@ -0,0 +1,11 @@
+{
+  "layer_idx": 8,
+  "model_dim": 4096,
+  "probe_dim": 64,
+  "num_layers": 24,
+  "num_heads": 4,
+  "ffn_multiplier": 4,
+  "dropout": 0.0,
+  "train_mse": 0.9838169354915619,
+  "eval_mse": 0.7481966165184717
+}