rttl's picture
Update README.md
18461c1
---
license: apache-2.0
language:
- en
model-index:
- name: rttl-ai/SentyBert
results:
- task:
type: task-classification
name: Text Classification
dataset:
type: sst2
name: sst2
config: default
split: validation
metrics:
- type: f1
value: 0.9992
name: F1 Macro
- type: accuracy
value: 0.9992
name: Accuracy
datasets:
- sst2
- sst
---
# rttl-ai/SentyBert
## Model Details
**Model Description:** This model is a fine-tune checkpoint of [bert-large-uncased](https://huggingface.co/bert-large-uncased), fine-tuned on SST-2.
This model reaches an accuracy of 99.92 on the dev set.
- **Developed by:** rttl-ai
- **Model Type:** Text Classification
- **Language(s):** English
- **License:** Apache-2.0
- **Resources for more information:**
- The model was pre-trained with task-adaptive pre-training [TAPT](https://arxiv.org/pdf/2004.10964.pdf) with an increased masking rate, no corruption strategy, and using WWM, following [this paper](https://aclanthology.org/2023.eacl-main.217.pdf)
- fine-tuned on sst with subtrees
- fine-tuned on sst2