Instructions to use leukas/DeBaby-fullcontr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use leukas/DeBaby-fullcontr with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="leukas/DeBaby-fullcontr")# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("leukas/DeBaby-fullcontr") model = AutoModelForMaskedLM.from_pretrained("leukas/DeBaby-fullcontr") - Notebooks
- Google Colab
- Kaggle
| { | |
| "_name_or_path": "/mounts/work/faeze/babylm/models/new-simcse-mlm-pretrain-256-checkpoint/checkpoint-0", | |
| "architectures": [ | |
| "DebertaV2ForMaskedLM" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 768, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 3072, | |
| "layer_norm_eps": 1e-07, | |
| "max_position_embeddings": 1024, | |
| "max_relative_positions": -1, | |
| "model_type": "deberta-v2", | |
| "norm_rel_ebd": "layer_norm", | |
| "num_attention_heads": 12, | |
| "num_hidden_layers": 12, | |
| "pad_token_id": 0, | |
| "pooler_dropout": 0, | |
| "pooler_hidden_act": "gelu", | |
| "pooler_hidden_size": 768, | |
| "pos_att_type": [ | |
| "p2c", | |
| "c2p" | |
| ], | |
| "position_biased_input": false, | |
| "position_buckets": 256, | |
| "relative_attention": true, | |
| "share_att_key": true, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.44.2", | |
| "type_vocab_size": 0, | |
| "vocab_size": 40000 | |
| } | |