We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 0c92f36 commit 9a8f29cCopy full SHA for 9a8f29c
open_diloco/configs/config_150m.json
@@ -1,13 +1,10 @@
1
-{
2
- "architectures": [
3
- "LlamaForCausalLM"
4
- ],
5
- "model_type": "llama",
6
- "hidden_size": 1024,
7
- "intermediate_size": 2688,
8
- "num_attention_heads": 16,
9
- "num_hidden_layers": 12,
10
- "use_cache": false,
11
- "rms_norm_eps": 1e-05
12
-}
13
+{
+ "name": "llama150m",
+ "n_embd": 1024,
+ "intermediate_size": 4096,
+ "n_head": 16,
+ "n_layer": 12,
+ "vocab_size": 32000,
+ "block_size": 1024
+}
open_diloco/configs/config_1b.json
@@ -1,9 +1,10 @@
{
- "name": "llama",
+ "name": "llama1b",
"n_embd": 2048,
"intermediate_size": 5632,
"n_head": 32,
"n_layer": 22,
"n_query_groups": 4,
- "vocab_size": 1024
}
0 commit comments