Update config.json
Browse files- config.json +4 -7
config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "tiiuae/falcon-40b",
|
| 3 |
"alibi": false,
|
| 4 |
"apply_residual_connection_post_layernorm": false,
|
| 5 |
"architectures": [
|
|
@@ -7,12 +6,9 @@
|
|
| 7 |
],
|
| 8 |
"attention_dropout": 0.0,
|
| 9 |
"auto_map": {
|
| 10 |
-
"AutoConfig": "
|
| 11 |
-
"AutoModel": "
|
| 12 |
-
"AutoModelForCausalLM": "
|
| 13 |
-
"AutoModelForQuestionAnswering": "tiiuae/falcon-40b--modelling_RW.RWForQuestionAnswering",
|
| 14 |
-
"AutoModelForSequenceClassification": "tiiuae/falcon-40b--modelling_RW.RWForSequenceClassification",
|
| 15 |
-
"AutoModelForTokenClassification": "tiiuae/falcon-40b--modelling_RW.RWForTokenClassification"
|
| 16 |
},
|
| 17 |
"bias": false,
|
| 18 |
"bos_token_id": 11,
|
|
@@ -22,6 +18,7 @@
|
|
| 22 |
"initializer_range": 0.02,
|
| 23 |
"layer_norm_epsilon": 1e-05,
|
| 24 |
"model_type": "RefinedWeb",
|
|
|
|
| 25 |
"n_head": 128,
|
| 26 |
"n_head_kv": 8,
|
| 27 |
"n_layer": 60,
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"alibi": false,
|
| 3 |
"apply_residual_connection_post_layernorm": false,
|
| 4 |
"architectures": [
|
|
|
|
| 6 |
],
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"auto_map": {
|
| 9 |
+
"AutoConfig": "configuration_RW.RWConfig",
|
| 10 |
+
"AutoModel": "modelling_RW.RWModel",
|
| 11 |
+
"AutoModelForCausalLM": "modelling_RW.RWForCausalLM",
|
|
|
|
|
|
|
|
|
|
| 12 |
},
|
| 13 |
"bias": false,
|
| 14 |
"bos_token_id": 11,
|
|
|
|
| 18 |
"initializer_range": 0.02,
|
| 19 |
"layer_norm_epsilon": 1e-05,
|
| 20 |
"model_type": "RefinedWeb",
|
| 21 |
+
"multi_query": true,
|
| 22 |
"n_head": 128,
|
| 23 |
"n_head_kv": 8,
|
| 24 |
"n_layer": 60,
|