Add architecture config (D512/L16/H8/FFN2048/SEQ512/VOCAB8192, weight-tied)
Browse files- config.json +15 -0
config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": ["TinyStoriesGPT"],
|
| 3 |
+
"d_model": 512,
|
| 4 |
+
"n_layers": 16,
|
| 5 |
+
"n_heads": 8,
|
| 6 |
+
"ffn_dim": 2048,
|
| 7 |
+
"vocab_size": 8192,
|
| 8 |
+
"max_seq_len": 512,
|
| 9 |
+
"tie_word_embeddings": true,
|
| 10 |
+
"norm": "rmsnorm",
|
| 11 |
+
"activation": "gelu",
|
| 12 |
+
"attn": "causal",
|
| 13 |
+
"dtype": "float32",
|
| 14 |
+
"n_params": 54804992
|
| 15 |
+
}
|