Text Classification
Transformers
ONNX
Safetensors
Rust
English
tool-use
function-calling
tool-calling
guardrails
verifier
shadow-mode
Instructions to use cowWhySo/toolcall-verifier-classifier-production with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cowWhySo/toolcall-verifier-classifier-production with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="cowWhySo/toolcall-verifier-classifier-production")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("cowWhySo/toolcall-verifier-classifier-production", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "schema_version": "toolcall-verifier-tokenization-diagnostics/v1", | |
| "base_model": "microsoft/deberta-v3-small", | |
| "run_profile": "high_vram_quality", | |
| "max_length": 1280, | |
| "sample_rows": 5000, | |
| "token_length_percentiles": { | |
| "0.5": 283.0, | |
| "0.75": 343.0, | |
| "0.9": 436.0, | |
| "0.95": 505.0, | |
| "0.99": 680.0200000000004 | |
| }, | |
| "candidate_marker_missing_rows": 0, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0, | |
| "candidate_payload_token_budget_percentiles": { | |
| "0.01": 1084.99, | |
| "0.05": 1177.0, | |
| "0.1": 1202.0, | |
| "0.5": 1232.0 | |
| }, | |
| "by_split": { | |
| "calibration": { | |
| "rows": 223, | |
| "token_length_percentiles": { | |
| "0.5": 286.0, | |
| "0.9": 455.8, | |
| "0.95": 514.0999999999998, | |
| "0.99": 718.7800000000001 | |
| }, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "test": { | |
| "rows": 473, | |
| "token_length_percentiles": { | |
| "0.5": 280.0, | |
| "0.9": 444.40000000000003, | |
| "0.95": 519.5999999999999, | |
| "0.99": 672.8399999999999 | |
| }, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "train": { | |
| "rows": 4072, | |
| "token_length_percentiles": { | |
| "0.5": 283.0, | |
| "0.9": 435.0, | |
| "0.95": 507.4499999999998, | |
| "0.99": 688.1599999999999 | |
| }, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "validation": { | |
| "rows": 232, | |
| "token_length_percentiles": { | |
| "0.5": 280.5, | |
| "0.9": 418.6, | |
| "0.95": 447.5999999999999, | |
| "0.99": 567.0599999999998 | |
| }, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| } | |
| }, | |
| "by_label": { | |
| "needs_clarification": { | |
| "rows": 7, | |
| "token_length_p95": 403.4, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "tool_not_needed": { | |
| "rows": 1552, | |
| "token_length_p95": 326.89999999999964, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "valid": { | |
| "rows": 1626, | |
| "token_length_p95": 527.5, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "wrong_arguments_semantic": { | |
| "rows": 1069, | |
| "token_length_p95": 566.0, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| }, | |
| "wrong_tool_semantic": { | |
| "rows": 746, | |
| "token_length_p95": 566.75, | |
| "candidate_marker_truncated_rows": 0, | |
| "candidate_marker_truncated_rate": 0.0 | |
| } | |
| }, | |
| "schema_marker_truncated_rows": 0, | |
| "schema_marker_truncated_rate": 0.0, | |
| "candidate_truncation_hard_fail_rate": 0.005 | |
| } |