Update config.json
Browse files- config.json +16 -2
config.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "ettin-decoder-32m",
|
| 3 |
"architectures": [
|
| 4 |
-
"
|
| 5 |
],
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
|
@@ -9,6 +9,7 @@
|
|
| 9 |
"classifier_activation": "gelu",
|
| 10 |
"classifier_bias": false,
|
| 11 |
"classifier_dropout": 0.0,
|
|
|
|
| 12 |
"cls_token_id": 50281,
|
| 13 |
"decoder_bias": true,
|
| 14 |
"deterministic_flash_attn": false,
|
|
@@ -46,5 +47,18 @@
|
|
| 46 |
"eos_token": "[SEP]",
|
| 47 |
"bos_token": "[CLS]",
|
| 48 |
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 49 |
-
"unk_token": "[UNK]"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "ettin-decoder-32m",
|
| 3 |
"architectures": [
|
| 4 |
+
"ModernBertDecoderForCausalLM"
|
| 5 |
],
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
|
|
|
| 9 |
"classifier_activation": "gelu",
|
| 10 |
"classifier_bias": false,
|
| 11 |
"classifier_dropout": 0.0,
|
| 12 |
+
"classifier_pooling": "mean",
|
| 13 |
"cls_token_id": 50281,
|
| 14 |
"decoder_bias": true,
|
| 15 |
"deterministic_flash_attn": false,
|
|
|
|
| 47 |
"eos_token": "[SEP]",
|
| 48 |
"bos_token": "[CLS]",
|
| 49 |
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 50 |
+
"unk_token": "[UNK]",
|
| 51 |
+
"layer_types": [
|
| 52 |
+
"full_attention",
|
| 53 |
+
"sliding_attention",
|
| 54 |
+
"sliding_attention",
|
| 55 |
+
"full_attention",
|
| 56 |
+
"sliding_attention",
|
| 57 |
+
"sliding_attention",
|
| 58 |
+
"full_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"full_attention"
|
| 62 |
+
],
|
| 63 |
+
"use_cache": true
|
| 64 |
}
|