Instructions to use openai/jukebox-5b-lyrics with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use openai/jukebox-5b-lyrics with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="openai/jukebox-5b-lyrics")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("openai/jukebox-5b-lyrics", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update config.json
#5
by ArthurZ HF Staff - opened
- config.json +3 -3
config.json
CHANGED
|
@@ -32,7 +32,7 @@
|
|
| 32 |
"alignment_head": 2,
|
| 33 |
"alignment_layer": 68,
|
| 34 |
"attention_multiplier": 0.25,
|
| 35 |
-
"attention_pattern": "
|
| 36 |
"attn_dropout": 0.0,
|
| 37 |
"attn_res_scale": false,
|
| 38 |
"blocks": 32,
|
|
@@ -132,7 +132,7 @@
|
|
| 132 |
"alignment_head": null,
|
| 133 |
"alignment_layer": null,
|
| 134 |
"attention_multiplier": 0.25,
|
| 135 |
-
"attention_pattern": "
|
| 136 |
"attn_dropout": 0,
|
| 137 |
"attn_res_scale": false,
|
| 138 |
"blocks": 64,
|
|
@@ -189,7 +189,7 @@
|
|
| 189 |
"alignment_head": null,
|
| 190 |
"alignment_layer": null,
|
| 191 |
"attention_multiplier": 0.25,
|
| 192 |
-
"attention_pattern": "
|
| 193 |
"attn_dropout": 0,
|
| 194 |
"attn_res_scale": false,
|
| 195 |
"blocks": 64,
|
|
|
|
| 32 |
"alignment_head": 2,
|
| 33 |
"alignment_layer": 68,
|
| 34 |
"attention_multiplier": 0.25,
|
| 35 |
+
"attention_pattern": "raw_column_previous_row_attention",
|
| 36 |
"attn_dropout": 0.0,
|
| 37 |
"attn_res_scale": false,
|
| 38 |
"blocks": 32,
|
|
|
|
| 132 |
"alignment_head": null,
|
| 133 |
"alignment_layer": null,
|
| 134 |
"attention_multiplier": 0.25,
|
| 135 |
+
"attention_pattern": "raw_column_previous_row_attention",
|
| 136 |
"attn_dropout": 0,
|
| 137 |
"attn_res_scale": false,
|
| 138 |
"blocks": 64,
|
|
|
|
| 189 |
"alignment_head": null,
|
| 190 |
"alignment_layer": null,
|
| 191 |
"attention_multiplier": 0.25,
|
| 192 |
+
"attention_pattern": "raw_column_previous_row_attention",
|
| 193 |
"attn_dropout": 0,
|
| 194 |
"attn_res_scale": false,
|
| 195 |
"blocks": 64,
|