-
EOS-254: v11 targeted data (lxc/-ct, stacks, search_apps recommend, container_logs, smb)
-
EOS-248 round2: v10 (regression fixes: pull-vs-install, +lxc-ct, list_docker_images, snapraid, abstain trim)
-
EOS-248: v9 dataset (list_docker_containers rename + 46 hard-case examples)
-
EOS-248: v9 dataset (list_docker_containers rename + 46 hard-case examples)
-
EOS-245: v9 contextual+guards, scoped top-16
-
EOS-243: v8 — per-example top-16 retrieved tool subsets (exemplar-enriched nomic-embed)
-
EOS-241: v6 — targeted classes from the EOS-240 failure inventory
-
EOS-236: v5 — casual paraphrases truncated at the tool call (rev1 canned-answer repetition taught answer-skipping)
-
EOS-236: casual paraphrase generator
-
EOS-236: q16 dataset rev — casual-register paraphrases + contrastive pairs (leak-checked vs eval)
-
Upload merged-dataset-v3.jsonl with huggingface_hub
-
Upload merged-dataset-v2.jsonl with huggingface_hub
-
Upload scripts/augment_final.py with huggingface_hub
-
Upload scripts/augment_no_tool.py with huggingface_hub
-
Upload scripts/augment_multiturn.py with huggingface_hub
-
Upload scripts/augment_grounding.py with huggingface_hub
-
Upload scripts/augment_errors.py with huggingface_hub
-
Upload scripts/augment_negatives.py with huggingface_hub
-
Upload scripts/generate_eval.py with huggingface_hub
-
Upload scripts/generate_dataset.py with huggingface_hub
-
Upload scripts/merge_and_quantize.py with huggingface_hub
-
Upload scripts/train.py with huggingface_hub
-
Upload scripts/train_functiongemma.py with huggingface_hub
-
Upload tools.json with huggingface_hub
-
Upload eval_dataset.jsonl with huggingface_hub
-
Upload no-tool-examples.jsonl with huggingface_hub
-
Upload nova-tool-calling-dataset.jsonl with huggingface_hub
-
Upload merged-dataset.jsonl with huggingface_hub
-
Upload README.md with huggingface_hub
-
initial commit