Update README.md
Browse files
README.md
CHANGED
@@ -50,14 +50,17 @@ More information needed
|
|
50 |
## Training procedure
|
51 |
|
52 |
training_config:
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
|
|
|
|
|
|
|
61 |
|
62 |
### Training hyperparameters
|
63 |
|
|
|
50 |
## Training procedure
|
51 |
|
52 |
training_config:
|
53 |
+
vocab_size: 30000
|
54 |
+
hidden_size: 256
|
55 |
+
intermediate_size: 512
|
56 |
+
num_hidden_layers: 4
|
57 |
+
num_attention_heads: 4
|
58 |
+
num_key_value_heads: 4
|
59 |
+
sliding_window: 4
|
60 |
+
max_position_embeddings: 1024
|
61 |
+
pad_token_id: 0
|
62 |
+
bos_token_id: 1
|
63 |
+
eos_token_id: 2
|
64 |
|
65 |
### Training hyperparameters
|
66 |
|