Upload Gemma2ForCausalLM/size=64/base/config.json with huggingface_hub
Browse files
Gemma2ForCausalLM/size=64/base/config.json
CHANGED
@@ -10,7 +10,7 @@
|
|
10 |
"cache_implementation": "hybrid",
|
11 |
"eos_token_id": 1,
|
12 |
"final_logit_softcapping": 30.0,
|
13 |
-
"head_dim":
|
14 |
"hidden_act": "gelu_pytorch_tanh",
|
15 |
"hidden_activation": "gelu_pytorch_tanh",
|
16 |
"hidden_size": 64,
|
@@ -22,7 +22,7 @@
|
|
22 |
"num_hidden_layers": 26,
|
23 |
"num_key_value_heads": 4,
|
24 |
"pad_token_id": 0,
|
25 |
-
"query_pre_attn_scalar":
|
26 |
"rms_norm_eps": 1e-06,
|
27 |
"rope_theta": 10000.0,
|
28 |
"sliding_window": 4096,
|
|
|
10 |
"cache_implementation": "hybrid",
|
11 |
"eos_token_id": 1,
|
12 |
"final_logit_softcapping": 30.0,
|
13 |
+
"head_dim": 64,
|
14 |
"hidden_act": "gelu_pytorch_tanh",
|
15 |
"hidden_activation": "gelu_pytorch_tanh",
|
16 |
"hidden_size": 64,
|
|
|
22 |
"num_hidden_layers": 26,
|
23 |
"num_key_value_heads": 4,
|
24 |
"pad_token_id": 0,
|
25 |
+
"query_pre_attn_scalar": 64,
|
26 |
"rms_norm_eps": 1e-06,
|
27 |
"rope_theta": 10000.0,
|
28 |
"sliding_window": 4096,
|