Edit model card

Metrics

  • loss: 0.9428
  • accuracy: 0.8265
  • precision: 0.8391
  • recall: 0.8265
  • precision_macro: 0.8497
  • recall_macro: 0.7568
  • macro_fpr: 0.0151
  • weighted_fpr: 0.0148
  • weighted_specificity: 0.9777
  • macro_specificity: 0.9869
  • weighted_sensitivity: 0.8265
  • macro_sensitivity: 0.7568
  • f1_micro: 0.8265
  • f1_macro: 0.7735
  • f1_weighted: 0.8266
  • runtime: 19.1857
  • samples_per_second: 67.2900
  • steps_per_second: 8.4440

legal-InLegal-merge-passthrough

legal-InLegal-merge-passthrough is a merge of the following models using mergekit:

🧩 Configuration

models:
  - model: xshubhamx/legal-bert-base-uncased
    # No parameters necessary for base model
  - model: xshubhamx/InLegalBERT
    parameters:
      density: 0.53
      weight: 1
merge_method: dare_ties
base_model: xshubhamx/legal-bert-base-uncased
parameters:
  int8_mask: true
dtype: bfloat16
Downloads last month
2
Safetensors
Model size
109M params
Tensor type
BF16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.