BRP Tiny-Transformers
Collection
Models for the 2024-Q4 BSc. Research Project: "Architectural Decisions for Language Modelling with Small Transformers". • 13 items • Updated • 1
# Load model directly
from transformers import AutoTokenizer, AutoModelForMaskedLM
tokenizer = AutoTokenizer.from_pretrained("AISE-TUDelft/Custom-Activations-BERT-SiLU")
model = AutoModelForMaskedLM.from_pretrained("AISE-TUDelft/Custom-Activations-BERT-SiLU")Quick Links
Basemodel: roBERTa
Configs: Vocab size: 10,000 Hidden size: 512 Max position embeddings: 512 Number of layers: 2 Number of heads: 4 Window size: 256 Intermediate-size: 1024
Results:
- Task: glue Score: 57.54 Confidence Interval: [57.15, 58.1]
- Task: blimp Score: 59.16 Confidence Interval: [58.75, 59.53]
- Downloads last month
- 3
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="AISE-TUDelft/Custom-Activations-BERT-SiLU")