diff --git a/examples/distillation/training_configs/distilbert-base-uncased.json b/examples/distillation/training_configs/distilbert-base-uncased.json
new file mode 100644
index 00000000..15d1e7fe
--- /dev/null
+++ b/examples/distillation/training_configs/distilbert-base-uncased.json
@@ -0,0 +1,15 @@
+{
+	"activation": "gelu",
+	"attention_dropout": 0.1,
+	"dim": 768,
+	"dropout": 0.1,
+	"hidden_dim": 3072,
+	"initializer_range": 0.02,
+	"max_position_embeddings": 512,
+	"n_heads": 12,
+	"n_layers": 6,
+	"sinusoidal_pos_embds": true,
+	"tie_weights_": true,
+	"vocab_size": 30522
+  }
+  
\ No newline at end of file
diff --git a/examples/distillation/training_configs/distilgpt2.json b/examples/distillation/training_configs/distilgpt2.json
new file mode 100644
index 00000000..8616e8e6
--- /dev/null
+++ b/examples/distillation/training_configs/distilgpt2.json
@@ -0,0 +1,10 @@
+{
+	"initializer_range": 0.02,
+	"layer_norm_epsilon": 0.00001,
+	"n_ctx": 1024,
+	"n_embd": 768,
+	"n_head": 12,
+	"n_layer": 6,
+	"n_positions": 1024,
+	"vocab_size": 50257
+}
\ No newline at end of file