update max tokens based on 30sec (max trained sample) * 6.5 toks/sec
Browse files- config.json +2 -1
- generation_config.json +3 -1
config.json
CHANGED
|
@@ -6,6 +6,7 @@
|
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"bos_token_id": 1,
|
|
|
|
| 9 |
"decoder_hidden_act": "silu",
|
| 10 |
"decoder_num_attention_heads": 8,
|
| 11 |
"decoder_num_hidden_layers": 8,
|
|
@@ -20,7 +21,7 @@
|
|
| 20 |
"initializer_range": 0.02,
|
| 21 |
"intermediate_size": 1664,
|
| 22 |
"is_encoder_decoder": true,
|
| 23 |
-
"max_position_embeddings":
|
| 24 |
"model_type": "moonshine",
|
| 25 |
"partial_rotary_factor": 0.62,
|
| 26 |
"rope_scaling": null,
|
|
|
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"bos_token_id": 1,
|
| 9 |
+
"pad_token_id": 2,
|
| 10 |
"decoder_hidden_act": "silu",
|
| 11 |
"decoder_num_attention_heads": 8,
|
| 12 |
"decoder_num_hidden_layers": 8,
|
|
|
|
| 21 |
"initializer_range": 0.02,
|
| 22 |
"intermediate_size": 1664,
|
| 23 |
"is_encoder_decoder": true,
|
| 24 |
+
"max_position_embeddings": 194,
|
| 25 |
"model_type": "moonshine",
|
| 26 |
"partial_rotary_factor": 0.62,
|
| 27 |
"rope_scaling": null,
|
generation_config.json
CHANGED
|
@@ -1,7 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 1,
|
|
|
|
| 4 |
"decoder_start_token_id": 1,
|
| 5 |
"eos_token_id": 2,
|
| 6 |
-
"transformers_version": "4.48.0.dev0"
|
|
|
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 1,
|
| 4 |
+
"pad_token_id": 2,
|
| 5 |
"decoder_start_token_id": 1,
|
| 6 |
"eos_token_id": 2,
|
| 7 |
+
"transformers_version": "4.48.0.dev0",
|
| 8 |
+
"max_length": 194
|
| 9 |
}
|