YAML Metadata Warning:The pipeline tag "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
TensorCode evidence-conditioned chatbot
Owned sequence encoder, relational slot workspace, and language decoder. Initialized
from google/flan-t5-small at 0fc9ddf78a1e988dac52e2dac162b0ede4fd74ab and fine-tuned using the explicit data
hashes and source IDs in evaluation.json. Schedule: 3 epochs; seed 7.
This checkpoint is a narrow evidence-conditioned QA demonstration, not evidence of general autonomous cognition. Foundation instruction behavior is inherited; training updates both foundation and workspace. Data preparation may supply oracle supporting passages: retrieval competence is not established by this experiment. The model does not independently verify its generated claims. Workspace bypass and zero-evidence ablations are reported separately; zero ablation alone does not show workspace utility.
Held-out exact match: 0.4688; token F1: 0.6015. Workspace-bypassed exact match: 0.4688. See evaluation.json for every prediction, before/after metrics, source provenance and limitations. Published weights contain no runtime conversation; training examples must be licensed and reviewed by their supplier.
from tensorcode.tools.chatbot import Chatbot
model = Chatbot.from_pretrained("PATH_OR_HUB_ID")
answer = model.generate_batch(["Question: ...\nEvidence: ..."])[0]
Model tree for jacob-valdez/tensorcode-chatbot-hotpot-001
Base model
google/flan-t5-small