train.jsonl

{"id": "compactlm-0453", "category": "extraction", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing retrieval augmentation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"retrieval augmentation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0577", "category": "extraction", "topic": "model deployment", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing model deployment and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model deployment\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0155", "category": "glossary", "topic": "training validation", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for training validation, including one practical consideration."}, {"role": "assistant", "content": "Training validation: using held-out examples to assess model behavior during development. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0513", "category": "glossary", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for retrieval augmentation, including one practical consideration."}, {"role": "assistant", "content": "Retrieval augmentation: supplementing generation with information retrieved from an external collection. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0036", "category": "glossary", "topic": "context windows", "messages": [{"role": "user", "content": "Write a short glossary entry for context windows, including one practical consideration."}, {"role": "assistant", "content": "Context windows: the limit on how many input and generated tokens a model can consider in one request. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0441", "category": "extraction", "topic": "GPU inference", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing GPU inference and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"GPU inference\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0074", "category": "rewriting", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing fine-tuning stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating fine-tuning and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0006", "category": "definition", "topic": "batching", "messages": [{"role": "user", "content": "Explain batching in one sentence for a beginner."}, {"role": "assistant", "content": "Batching means processing multiple requests together to improve hardware utilization."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0051", "category": "summarization", "topic": "latency optimization", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring latency optimization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot latency optimization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0185", "category": "rewriting", "topic": "quantization", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing quantization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating quantization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0113", "category": "evaluation", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving retrieval augmentation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for retrieval augmentation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0398", "category": "glossary", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for hallucination detection, including one practical consideration."}, {"role": "assistant", "content": "Hallucination detection: identifying generated claims that lack support or are incorrect. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0489", "category": "definition", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For technical documentation: Explain prompt evaluation in one sentence for a beginner."}, {"role": "assistant", "content": "Prompt evaluation means testing how model outputs change in response to prompts."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0109", "category": "evaluation", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving prompt evaluation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for prompt evaluation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0348", "category": "evaluation", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving dataset curation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for dataset curation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0491", "category": "definition", "topic": "latency optimization", "messages": [{"role": "user", "content": "For technical documentation: Explain latency optimization in one sentence for a beginner."}, {"role": "assistant", "content": "Latency optimization means reducing the delay between an input request and a generated response."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0246", "category": "definition", "topic": "batching", "messages": [{"role": "user", "content": "For a small research prototype: Explain batching in one sentence for a beginner."}, {"role": "assistant", "content": "Batching means processing multiple requests together to improve hardware utilization."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0335", "category": "extraction", "topic": "training validation", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing training validation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"training validation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0590", "category": "evaluation", "topic": "code generation", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving code generation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for code generation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0234", "category": "evaluation", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving fine-tuning."}, {"role": "assistant", "content": "1. Record the test setup and baseline for fine-tuning.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0511", "category": "glossary", "topic": "latency optimization", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for latency optimization, including one practical consideration."}, {"role": "assistant", "content": "Latency optimization: reducing the delay between an input request and a generated response. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0473", "category": "evaluation", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving retrieval augmentation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for retrieval augmentation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0358", "category": "evaluation", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving hallucination detection."}, {"role": "assistant", "content": "1. Record the test setup and baseline for hallucination detection.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0277", "category": "glossary", "topic": "model deployment", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for model deployment, including one practical consideration."}, {"role": "assistant", "content": "Model deployment: making a model available for use in an application or service. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0519", "category": "glossary", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for classifier calibration, including one practical consideration."}, {"role": "assistant", "content": "Classifier calibration: checking whether predicted probabilities reflect observed outcome frequencies. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0221", "category": "evaluation", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving GPU inference."}, {"role": "assistant", "content": "1. Record the test setup and baseline for GPU inference.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0318", "category": "rewriting", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing hallucination detection stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating hallucination detection and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0435", "category": "rewriting", "topic": "training validation", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing training validation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating training validation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0174", "category": "summarization", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring fine-tuning. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot fine-tuning, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0353", "category": "evaluation", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving retrieval augmentation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for retrieval augmentation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0526", "category": "summarization", "topic": "batching", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring batching. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot batching, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0382", "category": "glossary", "topic": "on-device AI", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for on-device AI, including one practical consideration."}, {"role": "assistant", "content": "On-device ai: executing AI models locally on a phone, laptop, or other device. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0381", "category": "glossary", "topic": "GPU inference", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for GPU inference, including one practical consideration."}, {"role": "assistant", "content": "Gpu inference: running a trained model on a graphics processor to produce predictions or outputs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0394", "category": "glossary", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for fine-tuning, including one practical consideration."}, {"role": "assistant", "content": "Fine-tuning: continuing model training on a smaller task- or domain-specific dataset. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0331", "category": "extraction", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing latency optimization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"latency optimization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0311", "category": "rewriting", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing latency optimization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating latency optimization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0232", "category": "evaluation", "topic": "memory usage", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving memory usage."}, {"role": "assistant", "content": "1. Record the test setup and baseline for memory usage.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0278", "category": "glossary", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for hallucination detection, including one practical consideration."}, {"role": "assistant", "content": "Hallucination detection: identifying generated claims that lack support or are incorrect. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0047", "category": "summarization", "topic": "model distillation", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring model distillation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model distillation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0293", "category": "summarization", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring retrieval augmentation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot retrieval augmentation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0004", "category": "definition", "topic": "embedding search", "messages": [{"role": "user", "content": "Explain embedding search in one sentence for a beginner."}, {"role": "assistant", "content": "Embedding search means retrieving items using similarity between numerical vector representations."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0531", "category": "summarization", "topic": "latency optimization", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring latency optimization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot latency optimization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0346", "category": "evaluation", "topic": "batching", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving batching."}, {"role": "assistant", "content": "1. Record the test setup and baseline for batching.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0315", "category": "rewriting", "topic": "training validation", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing training validation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating training validation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0199", "category": "rewriting", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing classifier calibration stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating classifier calibration and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0103", "category": "evaluation", "topic": "tokenization", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving tokenization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for tokenization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0132", "category": "definition", "topic": "memory usage", "messages": [{"role": "user", "content": "For a student project: Explain memory usage in one sentence for a beginner."}, {"role": "assistant", "content": "Memory usage means the amount of RAM or accelerator memory consumed during execution."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0600", "category": "evaluation", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving synthetic datasets."}, {"role": "assistant", "content": "1. Record the test setup and baseline for synthetic datasets.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0067", "category": "rewriting", "topic": "model distillation", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing model distillation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model distillation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0108", "category": "evaluation", "topic": "dataset curation", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving dataset curation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for dataset curation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0576", "category": "extraction", "topic": "context windows", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing context windows and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"context windows\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0073", "category": "rewriting", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing retrieval augmentation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating retrieval augmentation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0173", "category": "summarization", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring retrieval augmentation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot retrieval augmentation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0484", "category": "definition", "topic": "embedding search", "messages": [{"role": "user", "content": "For technical documentation: Explain embedding search in one sentence for a beginner."}, {"role": "assistant", "content": "Embedding search means retrieving items using similarity between numerical vector representations."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0583", "category": "evaluation", "topic": "tokenization", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving tokenization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for tokenization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0049", "category": "summarization", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring prompt evaluation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot prompt evaluation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0469", "category": "evaluation", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving prompt evaluation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for prompt evaluation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0153", "category": "glossary", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for retrieval augmentation, including one practical consideration."}, {"role": "assistant", "content": "Retrieval augmentation: supplementing generation with information retrieved from an external collection. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0026", "category": "glossary", "topic": "batching", "messages": [{"role": "user", "content": "Write a short glossary entry for batching, including one practical consideration."}, {"role": "assistant", "content": "Batching: processing multiple requests together to improve hardware utilization. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0258", "category": "definition", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a small research prototype: Explain hallucination detection in one sentence for a beginner."}, {"role": "assistant", "content": "Hallucination detection means identifying generated claims that lack support or are incorrect."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0271", "category": "glossary", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for latency optimization, including one practical consideration."}, {"role": "assistant", "content": "Latency optimization: reducing the delay between an input request and a generated response. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0081", "category": "extraction", "topic": "GPU inference", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing GPU inference and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"GPU inference\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0548", "category": "rewriting", "topic": "dataset curation", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing dataset curation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating dataset curation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0054", "category": "summarization", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring fine-tuning. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot fine-tuning, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0097", "category": "extraction", "topic": "model deployment", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing model deployment and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model deployment\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0459", "category": "extraction", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing classifier calibration and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"classifier calibration\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0167", "category": "summarization", "topic": "model distillation", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring model distillation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model distillation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0261", "category": "glossary", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for GPU inference, including one practical consideration."}, {"role": "assistant", "content": "Gpu inference: running a trained model on a graphics processor to produce predictions or outputs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0025", "category": "glossary", "topic": "quantization", "messages": [{"role": "user", "content": "Write a short glossary entry for quantization, including one practical consideration."}, {"role": "assistant", "content": "Quantization: representing model weights or calculations using lower numerical precision. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0332", "category": "extraction", "topic": "memory usage", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing memory usage and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"memory usage\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0057", "category": "summarization", "topic": "model deployment", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring model deployment. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model deployment, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0587", "category": "evaluation", "topic": "model distillation", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving model distillation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model distillation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0352", "category": "evaluation", "topic": "memory usage", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving memory usage."}, {"role": "assistant", "content": "1. Record the test setup and baseline for memory usage.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0250", "category": "definition", "topic": "code generation", "messages": [{"role": "user", "content": "For a small research prototype: Explain code generation in one sentence for a beginner."}, {"role": "assistant", "content": "Code generation means producing programming code from natural-language or other input."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0017", "category": "definition", "topic": "model deployment", "messages": [{"role": "user", "content": "Explain model deployment in one sentence for a beginner."}, {"role": "assistant", "content": "Model deployment means making a model available for use in an application or service."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0585", "category": "evaluation", "topic": "quantization", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving quantization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for quantization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0440", "category": "rewriting", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing synthetic datasets stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating synthetic datasets and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0528", "category": "summarization", "topic": "dataset curation", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring dataset curation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot dataset curation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0455", "category": "extraction", "topic": "training validation", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing training validation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"training validation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0431", "category": "rewriting", "topic": "latency optimization", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing latency optimization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating latency optimization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0106", "category": "evaluation", "topic": "batching", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving batching."}, {"role": "assistant", "content": "1. Record the test setup and baseline for batching.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0259", "category": "definition", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a small research prototype: Explain classifier calibration in one sentence for a beginner."}, {"role": "assistant", "content": "Classifier calibration means checking whether predicted probabilities reflect observed outcome frequencies."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0238", "category": "evaluation", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving hallucination detection."}, {"role": "assistant", "content": "1. Record the test setup and baseline for hallucination detection.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0593", "category": "evaluation", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving retrieval augmentation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for retrieval augmentation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0267", "category": "glossary", "topic": "model distillation", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for model distillation, including one practical consideration."}, {"role": "assistant", "content": "Model distillation: training a smaller student model using guidance from a stronger teacher. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0295", "category": "summarization", "topic": "training validation", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring training validation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot training validation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0076", "category": "rewriting", "topic": "context windows", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing context windows stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating context windows and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0086", "category": "extraction", "topic": "batching", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing batching and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"batching\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0060", "category": "summarization", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring synthetic datasets. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot synthetic datasets, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0533", "category": "summarization", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring retrieval augmentation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot retrieval augmentation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0434", "category": "rewriting", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing fine-tuning stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating fine-tuning and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0321", "category": "extraction", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing GPU inference and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"GPU inference\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0066", "category": "rewriting", "topic": "batching", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing batching stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating batching and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0001", "category": "definition", "topic": "GPU inference", "messages": [{"role": "user", "content": "Explain GPU inference in one sentence for a beginner."}, {"role": "assistant", "content": "Gpu inference means running a trained model on a graphics processor to produce predictions or outputs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0288", "category": "summarization", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring dataset curation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot dataset curation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0043", "category": "summarization", "topic": "tokenization", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring tokenization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot tokenization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0244", "category": "definition", "topic": "embedding search", "messages": [{"role": "user", "content": "For a small research prototype: Explain embedding search in one sentence for a beginner."}, {"role": "assistant", "content": "Embedding search means retrieving items using similarity between numerical vector representations."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0574", "category": "extraction", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing fine-tuning and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"fine-tuning\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0329", "category": "extraction", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing prompt evaluation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"prompt evaluation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0543", "category": "rewriting", "topic": "tokenization", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing tokenization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating tokenization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0186", "category": "rewriting", "topic": "batching", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing batching stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating batching and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0538", "category": "summarization", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring hallucination detection. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot hallucination detection, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0260", "category": "definition", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a small research prototype: Explain synthetic datasets in one sentence for a beginner."}, {"role": "assistant", "content": "Synthetic datasets means collections of artificially generated examples rather than directly observed samples."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0403", "category": "summarization", "topic": "tokenization", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring tokenization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot tokenization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0012", "category": "definition", "topic": "memory usage", "messages": [{"role": "user", "content": "Explain memory usage in one sentence for a beginner."}, {"role": "assistant", "content": "Memory usage means the amount of RAM or accelerator memory consumed during execution."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0030", "category": "glossary", "topic": "code generation", "messages": [{"role": "user", "content": "Write a short glossary entry for code generation, including one practical consideration."}, {"role": "assistant", "content": "Code generation: producing programming code from natural-language or other input. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0194", "category": "rewriting", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing fine-tuning stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating fine-tuning and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0105", "category": "evaluation", "topic": "quantization", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving quantization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for quantization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0290", "category": "summarization", "topic": "code generation", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring code generation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot code generation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0504", "category": "glossary", "topic": "embedding search", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for embedding search, including one practical consideration."}, {"role": "assistant", "content": "Embedding search: retrieving items using similarity between numerical vector representations. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0218", "category": "extraction", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing hallucination detection and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"hallucination detection\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0558", "category": "rewriting", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing hallucination detection stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating hallucination detection and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0557", "category": "rewriting", "topic": "model deployment", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing model deployment stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model deployment and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0137", "category": "definition", "topic": "model deployment", "messages": [{"role": "user", "content": "For a student project: Explain model deployment in one sentence for a beginner."}, {"role": "assistant", "content": "Model deployment means making a model available for use in an application or service."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0149", "category": "glossary", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for prompt evaluation, including one practical consideration."}, {"role": "assistant", "content": "Prompt evaluation: testing how model outputs change in response to prompts. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0210", "category": "extraction", "topic": "code generation", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing code generation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"code generation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0563", "category": "extraction", "topic": "tokenization", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing tokenization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"tokenization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0003", "category": "definition", "topic": "tokenization", "messages": [{"role": "user", "content": "Explain tokenization in one sentence for a beginner."}, {"role": "assistant", "content": "Tokenization means turning text into smaller units represented by numeric token IDs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0420", "category": "summarization", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring synthetic datasets. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot synthetic datasets, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0298", "category": "summarization", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring hallucination detection. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot hallucination detection, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0039", "category": "glossary", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Write a short glossary entry for classifier calibration, including one practical consideration."}, {"role": "assistant", "content": "Classifier calibration: checking whether predicted probabilities reflect observed outcome frequencies. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0439", "category": "rewriting", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing classifier calibration stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating classifier calibration and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0275", "category": "glossary", "topic": "training validation", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for training validation, including one practical consideration."}, {"role": "assistant", "content": "Training validation: using held-out examples to assess model behavior during development. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0552", "category": "rewriting", "topic": "memory usage", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing memory usage stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating memory usage and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0392", "category": "glossary", "topic": "memory usage", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for memory usage, including one practical consideration."}, {"role": "assistant", "content": "Memory usage: the amount of RAM or accelerator memory consumed during execution. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0597", "category": "evaluation", "topic": "model deployment", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving model deployment."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model deployment.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0481", "category": "definition", "topic": "GPU inference", "messages": [{"role": "user", "content": "For technical documentation: Explain GPU inference in one sentence for a beginner."}, {"role": "assistant", "content": "Gpu inference means running a trained model on a graphics processor to produce predictions or outputs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0233", "category": "evaluation", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving retrieval augmentation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for retrieval augmentation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0115", "category": "evaluation", "topic": "training validation", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving training validation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for training validation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0480", "category": "evaluation", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving synthetic datasets."}, {"role": "assistant", "content": "1. Record the test setup and baseline for synthetic datasets.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0128", "category": "definition", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a student project: Explain dataset curation in one sentence for a beginner."}, {"role": "assistant", "content": "Dataset curation means selecting, cleaning, and documenting examples used for training or evaluation."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0107", "category": "evaluation", "topic": "model distillation", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving model distillation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model distillation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0228", "category": "evaluation", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving dataset curation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for dataset curation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0432", "category": "rewriting", "topic": "memory usage", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing memory usage stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating memory usage and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0280", "category": "glossary", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for synthetic datasets, including one practical consideration."}, {"role": "assistant", "content": "Synthetic datasets: collections of artificially generated examples rather than directly observed samples. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0452", "category": "extraction", "topic": "memory usage", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing memory usage and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"memory usage\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0056", "category": "summarization", "topic": "context windows", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring context windows. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot context windows, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0418", "category": "summarization", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring hallucination detection. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot hallucination detection, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0240", "category": "evaluation", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving synthetic datasets."}, {"role": "assistant", "content": "1. Record the test setup and baseline for synthetic datasets.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0340", "category": "extraction", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing synthetic datasets and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"synthetic datasets\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0488", "category": "definition", "topic": "dataset curation", "messages": [{"role": "user", "content": "For technical documentation: Explain dataset curation in one sentence for a beginner."}, {"role": "assistant", "content": "Dataset curation means selecting, cleaning, and documenting examples used for training or evaluation."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0062", "category": "rewriting", "topic": "on-device AI", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing on-device AI stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating on-device AI and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0415", "category": "summarization", "topic": "training validation", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring training validation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot training validation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0456", "category": "extraction", "topic": "context windows", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing context windows and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"context windows\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0018", "category": "definition", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Explain hallucination detection in one sentence for a beginner."}, {"role": "assistant", "content": "Hallucination detection means identifying generated claims that lack support or are incorrect."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0196", "category": "rewriting", "topic": "context windows", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing context windows stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating context windows and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0239", "category": "evaluation", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving classifier calibration."}, {"role": "assistant", "content": "1. Record the test setup and baseline for classifier calibration.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0510", "category": "glossary", "topic": "code generation", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for code generation, including one practical consideration."}, {"role": "assistant", "content": "Code generation: producing programming code from natural-language or other input. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0339", "category": "extraction", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing classifier calibration and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"classifier calibration\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0265", "category": "glossary", "topic": "quantization", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for quantization, including one practical consideration."}, {"role": "assistant", "content": "Quantization: representing model weights or calculations using lower numerical precision. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0462", "category": "evaluation", "topic": "on-device AI", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving on-device AI."}, {"role": "assistant", "content": "1. Record the test setup and baseline for on-device AI.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0556", "category": "rewriting", "topic": "context windows", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing context windows stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating context windows and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0072", "category": "rewriting", "topic": "memory usage", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing memory usage stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating memory usage and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0467", "category": "evaluation", "topic": "model distillation", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving model distillation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model distillation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0324", "category": "extraction", "topic": "embedding search", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing embedding search and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"embedding search\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0360", "category": "evaluation", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving synthetic datasets."}, {"role": "assistant", "content": "1. Record the test setup and baseline for synthetic datasets.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0204", "category": "extraction", "topic": "embedding search", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing embedding search and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"embedding search\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0539", "category": "summarization", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring classifier calibration. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot classifier calibration, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0534", "category": "summarization", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring fine-tuning. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot fine-tuning, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0570", "category": "extraction", "topic": "code generation", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing code generation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"code generation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0317", "category": "rewriting", "topic": "model deployment", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing model deployment stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model deployment and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0449", "category": "extraction", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing prompt evaluation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"prompt evaluation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0184", "category": "rewriting", "topic": "embedding search", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing embedding search stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating embedding search and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0002", "category": "definition", "topic": "on-device AI", "messages": [{"role": "user", "content": "Explain on-device AI in one sentence for a beginner."}, {"role": "assistant", "content": "On-device ai means executing AI models locally on a phone, laptop, or other device."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0282", "category": "summarization", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring on-device AI. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot on-device AI, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0090", "category": "extraction", "topic": "code generation", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing code generation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"code generation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0536", "category": "summarization", "topic": "context windows", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring context windows. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot context windows, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0038", "category": "glossary", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Write a short glossary entry for hallucination detection, including one practical consideration."}, {"role": "assistant", "content": "Hallucination detection: identifying generated claims that lack support or are incorrect. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0065", "category": "rewriting", "topic": "quantization", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing quantization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating quantization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0357", "category": "evaluation", "topic": "model deployment", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving model deployment."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model deployment.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0345", "category": "evaluation", "topic": "quantization", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving quantization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for quantization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0064", "category": "rewriting", "topic": "embedding search", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing embedding search stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating embedding search and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0084", "category": "extraction", "topic": "embedding search", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing embedding search and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"embedding search\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0069", "category": "rewriting", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing prompt evaluation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating prompt evaluation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0466", "category": "evaluation", "topic": "batching", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving batching."}, {"role": "assistant", "content": "1. Record the test setup and baseline for batching.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0351", "category": "evaluation", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving latency optimization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for latency optimization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0170", "category": "summarization", "topic": "code generation", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring code generation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot code generation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0296", "category": "summarization", "topic": "context windows", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring context windows. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot context windows, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0501", "category": "glossary", "topic": "GPU inference", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for GPU inference, including one practical consideration."}, {"role": "assistant", "content": "Gpu inference: running a trained model on a graphics processor to produce predictions or outputs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0058", "category": "summarization", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring hallucination detection. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot hallucination detection, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0135", "category": "definition", "topic": "training validation", "messages": [{"role": "user", "content": "For a student project: Explain training validation in one sentence for a beginner."}, {"role": "assistant", "content": "Training validation means using held-out examples to assess model behavior during development."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0527", "category": "summarization", "topic": "model distillation", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring model distillation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model distillation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0512", "category": "glossary", "topic": "memory usage", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for memory usage, including one practical consideration."}, {"role": "assistant", "content": "Memory usage: the amount of RAM or accelerator memory consumed during execution. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0365", "category": "definition", "topic": "quantization", "messages": [{"role": "user", "content": "Answer concisely: Explain quantization in one sentence for a beginner."}, {"role": "assistant", "content": "Quantization means representing model weights or calculations using lower numerical precision."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0125", "category": "definition", "topic": "quantization", "messages": [{"role": "user", "content": "For a student project: Explain quantization in one sentence for a beginner."}, {"role": "assistant", "content": "Quantization means representing model weights or calculations using lower numerical precision."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0245", "category": "definition", "topic": "quantization", "messages": [{"role": "user", "content": "For a small research prototype: Explain quantization in one sentence for a beginner."}, {"role": "assistant", "content": "Quantization means representing model weights or calculations using lower numerical precision."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0416", "category": "summarization", "topic": "context windows", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring context windows. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot context windows, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0463", "category": "evaluation", "topic": "tokenization", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving tokenization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for tokenization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0376", "category": "definition", "topic": "context windows", "messages": [{"role": "user", "content": "Answer concisely: Explain context windows in one sentence for a beginner."}, {"role": "assistant", "content": "Context windows means the limit on how many input and generated tokens a model can consider in one request."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0053", "category": "summarization", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring retrieval augmentation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot retrieval augmentation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0575", "category": "extraction", "topic": "training validation", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing training validation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"training validation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0502", "category": "glossary", "topic": "on-device AI", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for on-device AI, including one practical consideration."}, {"role": "assistant", "content": "On-device ai: executing AI models locally on a phone, laptop, or other device. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0176", "category": "summarization", "topic": "context windows", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring context windows. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot context windows, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0532", "category": "summarization", "topic": "memory usage", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring memory usage. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot memory usage, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0468", "category": "evaluation", "topic": "dataset curation", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving dataset curation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for dataset curation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0227", "category": "evaluation", "topic": "model distillation", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving model distillation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model distillation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0134", "category": "definition", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a student project: Explain fine-tuning in one sentence for a beginner."}, {"role": "assistant", "content": "Fine-tuning means continuing model training on a smaller task- or domain-specific dataset."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0598", "category": "evaluation", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving hallucination detection."}, {"role": "assistant", "content": "1. Record the test setup and baseline for hallucination detection.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0505", "category": "glossary", "topic": "quantization", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for quantization, including one practical consideration."}, {"role": "assistant", "content": "Quantization: representing model weights or calculations using lower numerical precision. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0189", "category": "rewriting", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing prompt evaluation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating prompt evaluation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0442", "category": "extraction", "topic": "on-device AI", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing on-device AI and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"on-device AI\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0052", "category": "summarization", "topic": "memory usage", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring memory usage. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot memory usage, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0419", "category": "summarization", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring classifier calibration. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot classifier calibration, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0451", "category": "extraction", "topic": "latency optimization", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing latency optimization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"latency optimization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0540", "category": "summarization", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring synthetic datasets. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot synthetic datasets, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0359", "category": "evaluation", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving classifier calibration."}, {"role": "assistant", "content": "1. Record the test setup and baseline for classifier calibration.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0138", "category": "definition", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a student project: Explain hallucination detection in one sentence for a beginner."}, {"role": "assistant", "content": "Hallucination detection means identifying generated claims that lack support or are incorrect."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0201", "category": "extraction", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing GPU inference and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"GPU inference\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0223", "category": "evaluation", "topic": "tokenization", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving tokenization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for tokenization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0399", "category": "glossary", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for classifier calibration, including one practical consideration."}, {"role": "assistant", "content": "Classifier calibration: checking whether predicted probabilities reflect observed outcome frequencies. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0269", "category": "glossary", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for prompt evaluation, including one practical consideration."}, {"role": "assistant", "content": "Prompt evaluation: testing how model outputs change in response to prompts. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0203", "category": "extraction", "topic": "tokenization", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing tokenization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"tokenization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0599", "category": "evaluation", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving classifier calibration."}, {"role": "assistant", "content": "1. Record the test setup and baseline for classifier calibration.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0040", "category": "glossary", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Write a short glossary entry for synthetic datasets, including one practical consideration."}, {"role": "assistant", "content": "Synthetic datasets: collections of artificially generated examples rather than directly observed samples. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0308", "category": "rewriting", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing dataset curation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating dataset curation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0070", "category": "rewriting", "topic": "code generation", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing code generation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating code generation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0312", "category": "rewriting", "topic": "memory usage", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing memory usage stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating memory usage and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0111", "category": "evaluation", "topic": "latency optimization", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving latency optimization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for latency optimization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0495", "category": "definition", "topic": "training validation", "messages": [{"role": "user", "content": "For technical documentation: Explain training validation in one sentence for a beginner."}, {"role": "assistant", "content": "Training validation means using held-out examples to assess model behavior during development."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0071", "category": "rewriting", "topic": "latency optimization", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing latency optimization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating latency optimization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0514", "category": "glossary", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for fine-tuning, including one practical consideration."}, {"role": "assistant", "content": "Fine-tuning: continuing model training on a smaller task- or domain-specific dataset. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0175", "category": "summarization", "topic": "training validation", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring training validation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot training validation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0428", "category": "rewriting", "topic": "dataset curation", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing dataset curation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating dataset curation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0005", "category": "definition", "topic": "quantization", "messages": [{"role": "user", "content": "Explain quantization in one sentence for a beginner."}, {"role": "assistant", "content": "Quantization means representing model weights or calculations using lower numerical precision."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0364", "category": "definition", "topic": "embedding search", "messages": [{"role": "user", "content": "Answer concisely: Explain embedding search in one sentence for a beginner."}, {"role": "assistant", "content": "Embedding search means retrieving items using similarity between numerical vector representations."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0529", "category": "summarization", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring prompt evaluation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot prompt evaluation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0179", "category": "summarization", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring classifier calibration. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot classifier calibration, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0433", "category": "rewriting", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing retrieval augmentation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating retrieval augmentation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0209", "category": "extraction", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing prompt evaluation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"prompt evaluation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0024", "category": "glossary", "topic": "embedding search", "messages": [{"role": "user", "content": "Write a short glossary entry for embedding search, including one practical consideration."}, {"role": "assistant", "content": "Embedding search: retrieving items using similarity between numerical vector representations. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0177", "category": "summarization", "topic": "model deployment", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring model deployment. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model deployment, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0413", "category": "summarization", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring retrieval augmentation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot retrieval augmentation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0299", "category": "summarization", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring classifier calibration. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot classifier calibration, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0224", "category": "evaluation", "topic": "embedding search", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving embedding search."}, {"role": "assistant", "content": "1. Record the test setup and baseline for embedding search.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0230", "category": "evaluation", "topic": "code generation", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving code generation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for code generation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0596", "category": "evaluation", "topic": "context windows", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving context windows."}, {"role": "assistant", "content": "1. Record the test setup and baseline for context windows.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0537", "category": "summarization", "topic": "model deployment", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring model deployment. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model deployment, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0304", "category": "rewriting", "topic": "embedding search", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing embedding search stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating embedding search and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0559", "category": "rewriting", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing classifier calibration stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating classifier calibration and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0483", "category": "definition", "topic": "tokenization", "messages": [{"role": "user", "content": "For technical documentation: Explain tokenization in one sentence for a beginner."}, {"role": "assistant", "content": "Tokenization means turning text into smaller units represented by numeric token IDs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0118", "category": "evaluation", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving hallucination detection."}, {"role": "assistant", "content": "1. Record the test setup and baseline for hallucination detection.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0380", "category": "definition", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Answer concisely: Explain synthetic datasets in one sentence for a beginner."}, {"role": "assistant", "content": "Synthetic datasets means collections of artificially generated examples rather than directly observed samples."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0163", "category": "summarization", "topic": "tokenization", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring tokenization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot tokenization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0031", "category": "glossary", "topic": "latency optimization", "messages": [{"role": "user", "content": "Write a short glossary entry for latency optimization, including one practical consideration."}, {"role": "assistant", "content": "Latency optimization: reducing the delay between an input request and a generated response. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0560", "category": "rewriting", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing synthetic datasets stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating synthetic datasets and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0310", "category": "rewriting", "topic": "code generation", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing code generation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating code generation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0368", "category": "definition", "topic": "dataset curation", "messages": [{"role": "user", "content": "Answer concisely: Explain dataset curation in one sentence for a beginner."}, {"role": "assistant", "content": "Dataset curation means selecting, cleaning, and documenting examples used for training or evaluation."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0160", "category": "glossary", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for synthetic datasets, including one practical consideration."}, {"role": "assistant", "content": "Synthetic datasets: collections of artificially generated examples rather than directly observed samples. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0438", "category": "rewriting", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing hallucination detection stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating hallucination detection and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0220", "category": "extraction", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing synthetic datasets and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"synthetic datasets\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0121", "category": "definition", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a student project: Explain GPU inference in one sentence for a beginner."}, {"role": "assistant", "content": "Gpu inference means running a trained model on a graphics processor to produce predictions or outputs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0247", "category": "definition", "topic": "model distillation", "messages": [{"role": "user", "content": "For a small research prototype: Explain model distillation in one sentence for a beginner."}, {"role": "assistant", "content": "Model distillation means training a smaller student model using guidance from a stronger teacher."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0427", "category": "rewriting", "topic": "model distillation", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing model distillation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model distillation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0285", "category": "summarization", "topic": "quantization", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring quantization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot quantization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0530", "category": "summarization", "topic": "code generation", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring code generation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot code generation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0320", "category": "rewriting", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing synthetic datasets stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating synthetic datasets and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0216", "category": "extraction", "topic": "context windows", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing context windows and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"context windows\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0549", "category": "rewriting", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing prompt evaluation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating prompt evaluation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0591", "category": "evaluation", "topic": "latency optimization", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving latency optimization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for latency optimization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0011", "category": "definition", "topic": "latency optimization", "messages": [{"role": "user", "content": "Explain latency optimization in one sentence for a beginner."}, {"role": "assistant", "content": "Latency optimization means reducing the delay between an input request and a generated response."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0328", "category": "extraction", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing dataset curation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"dataset curation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0152", "category": "glossary", "topic": "memory usage", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for memory usage, including one practical consideration."}, {"role": "assistant", "content": "Memory usage: the amount of RAM or accelerator memory consumed during execution. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0032", "category": "glossary", "topic": "memory usage", "messages": [{"role": "user", "content": "Write a short glossary entry for memory usage, including one practical consideration."}, {"role": "assistant", "content": "Memory usage: the amount of RAM or accelerator memory consumed during execution. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0104", "category": "evaluation", "topic": "embedding search", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving embedding search."}, {"role": "assistant", "content": "1. Record the test setup and baseline for embedding search.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0020", "category": "definition", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Explain synthetic datasets in one sentence for a beginner."}, {"role": "assistant", "content": "Synthetic datasets means collections of artificially generated examples rather than directly observed samples."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0383", "category": "glossary", "topic": "tokenization", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for tokenization, including one practical consideration."}, {"role": "assistant", "content": "Tokenization: turning text into smaller units represented by numeric token IDs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0215", "category": "extraction", "topic": "training validation", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing training validation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"training validation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0562", "category": "extraction", "topic": "on-device AI", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing on-device AI and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"on-device AI\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0213", "category": "extraction", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing retrieval augmentation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"retrieval augmentation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0300", "category": "summarization", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring synthetic datasets. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot synthetic datasets, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0371", "category": "definition", "topic": "latency optimization", "messages": [{"role": "user", "content": "Answer concisely: Explain latency optimization in one sentence for a beginner."}, {"role": "assistant", "content": "Latency optimization means reducing the delay between an input request and a generated response."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0192", "category": "rewriting", "topic": "memory usage", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing memory usage stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating memory usage and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0007", "category": "definition", "topic": "model distillation", "messages": [{"role": "user", "content": "Explain model distillation in one sentence for a beginner."}, {"role": "assistant", "content": "Model distillation means training a smaller student model using guidance from a stronger teacher."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0208", "category": "extraction", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing dataset curation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"dataset curation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0485", "category": "definition", "topic": "quantization", "messages": [{"role": "user", "content": "For technical documentation: Explain quantization in one sentence for a beginner."}, {"role": "assistant", "content": "Quantization means representing model weights or calculations using lower numerical precision."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0157", "category": "glossary", "topic": "model deployment", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for model deployment, including one practical consideration."}, {"role": "assistant", "content": "Model deployment: making a model available for use in an application or service. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0322", "category": "extraction", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing on-device AI and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"on-device AI\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0337", "category": "extraction", "topic": "model deployment", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing model deployment and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model deployment\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0048", "category": "summarization", "topic": "dataset curation", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring dataset curation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot dataset curation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0147", "category": "glossary", "topic": "model distillation", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for model distillation, including one practical consideration."}, {"role": "assistant", "content": "Model distillation: training a smaller student model using guidance from a stronger teacher. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0478", "category": "evaluation", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving hallucination detection."}, {"role": "assistant", "content": "1. Record the test setup and baseline for hallucination detection.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0408", "category": "summarization", "topic": "dataset curation", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring dataset curation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot dataset curation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0572", "category": "extraction", "topic": "memory usage", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing memory usage and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"memory usage\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0541", "category": "rewriting", "topic": "GPU inference", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing GPU inference stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating GPU inference and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0145", "category": "glossary", "topic": "quantization", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for quantization, including one practical consideration."}, {"role": "assistant", "content": "Quantization: representing model weights or calculations using lower numerical precision. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0165", "category": "summarization", "topic": "quantization", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring quantization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot quantization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0521", "category": "summarization", "topic": "GPU inference", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring GPU inference. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot GPU inference, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0401", "category": "summarization", "topic": "GPU inference", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring GPU inference. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot GPU inference, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0268", "category": "glossary", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for dataset curation, including one practical consideration."}, {"role": "assistant", "content": "Dataset curation: selecting, cleaning, and documenting examples used for training or evaluation. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0306", "category": "rewriting", "topic": "batching", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing batching stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating batching and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0522", "category": "summarization", "topic": "on-device AI", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring on-device AI. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot on-device AI, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0126", "category": "definition", "topic": "batching", "messages": [{"role": "user", "content": "For a student project: Explain batching in one sentence for a beginner."}, {"role": "assistant", "content": "Batching means processing multiple requests together to improve hardware utilization."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0325", "category": "extraction", "topic": "quantization", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing quantization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"quantization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0294", "category": "summarization", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring fine-tuning. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot fine-tuning, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0140", "category": "definition", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a student project: Explain synthetic datasets in one sentence for a beginner."}, {"role": "assistant", "content": "Synthetic datasets means collections of artificially generated examples rather than directly observed samples."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0564", "category": "extraction", "topic": "embedding search", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing embedding search and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"embedding search\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0130", "category": "definition", "topic": "code generation", "messages": [{"role": "user", "content": "For a student project: Explain code generation in one sentence for a beginner."}, {"role": "assistant", "content": "Code generation means producing programming code from natural-language or other input."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0411", "category": "summarization", "topic": "latency optimization", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring latency optimization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot latency optimization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0476", "category": "evaluation", "topic": "context windows", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving context windows."}, {"role": "assistant", "content": "1. Record the test setup and baseline for context windows.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0374", "category": "definition", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Answer concisely: Explain fine-tuning in one sentence for a beginner."}, {"role": "assistant", "content": "Fine-tuning means continuing model training on a smaller task- or domain-specific dataset."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0507", "category": "glossary", "topic": "model distillation", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for model distillation, including one practical consideration."}, {"role": "assistant", "content": "Model distillation: training a smaller student model using guidance from a stronger teacher. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0087", "category": "extraction", "topic": "model distillation", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing model distillation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model distillation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0178", "category": "summarization", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring hallucination detection. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot hallucination detection, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0141", "category": "glossary", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for GPU inference, including one practical consideration."}, {"role": "assistant", "content": "Gpu inference: running a trained model on a graphics processor to produce predictions or outputs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0361", "category": "definition", "topic": "GPU inference", "messages": [{"role": "user", "content": "Answer concisely: Explain GPU inference in one sentence for a beginner."}, {"role": "assistant", "content": "Gpu inference means running a trained model on a graphics processor to produce predictions or outputs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0479", "category": "evaluation", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving classifier calibration."}, {"role": "assistant", "content": "1. Record the test setup and baseline for classifier calibration.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0378", "category": "definition", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Answer concisely: Explain hallucination detection in one sentence for a beginner."}, {"role": "assistant", "content": "Hallucination detection means identifying generated claims that lack support or are incorrect."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0400", "category": "glossary", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for synthetic datasets, including one practical consideration."}, {"role": "assistant", "content": "Synthetic datasets: collections of artificially generated examples rather than directly observed samples. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0477", "category": "evaluation", "topic": "model deployment", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving model deployment."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model deployment.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0516", "category": "glossary", "topic": "context windows", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for context windows, including one practical consideration."}, {"role": "assistant", "content": "Context windows: the limit on how many input and generated tokens a model can consider in one request. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0561", "category": "extraction", "topic": "GPU inference", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing GPU inference and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"GPU inference\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0100", "category": "extraction", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing synthetic datasets and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"synthetic datasets\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0444", "category": "extraction", "topic": "embedding search", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing embedding search and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"embedding search\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0389", "category": "glossary", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for prompt evaluation, including one practical consideration."}, {"role": "assistant", "content": "Prompt evaluation: testing how model outputs change in response to prompts. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0161", "category": "summarization", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring GPU inference. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot GPU inference, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0437", "category": "rewriting", "topic": "model deployment", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing model deployment stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model deployment and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0162", "category": "summarization", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring on-device AI. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot on-device AI, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0008", "category": "definition", "topic": "dataset curation", "messages": [{"role": "user", "content": "Explain dataset curation in one sentence for a beginner."}, {"role": "assistant", "content": "Dataset curation means selecting, cleaning, and documenting examples used for training or evaluation."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0579", "category": "extraction", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing classifier calibration and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"classifier calibration\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0046", "category": "summarization", "topic": "batching", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring batching. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot batching, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0082", "category": "extraction", "topic": "on-device AI", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing on-device AI and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"on-device AI\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0136", "category": "definition", "topic": "context windows", "messages": [{"role": "user", "content": "For a student project: Explain context windows in one sentence for a beginner."}, {"role": "assistant", "content": "Context windows means the limit on how many input and generated tokens a model can consider in one request."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0386", "category": "glossary", "topic": "batching", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for batching, including one practical consideration."}, {"role": "assistant", "content": "Batching: processing multiple requests together to improve hardware utilization. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0235", "category": "evaluation", "topic": "training validation", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving training validation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for training validation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0083", "category": "extraction", "topic": "tokenization", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing tokenization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"tokenization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0180", "category": "summarization", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring synthetic datasets. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot synthetic datasets, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0344", "category": "evaluation", "topic": "embedding search", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving embedding search."}, {"role": "assistant", "content": "1. Record the test setup and baseline for embedding search.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0349", "category": "evaluation", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving prompt evaluation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for prompt evaluation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0407", "category": "summarization", "topic": "model distillation", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring model distillation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model distillation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0044", "category": "summarization", "topic": "embedding search", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring embedding search. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot embedding search, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0460", "category": "extraction", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing synthetic datasets and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"synthetic datasets\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0550", "category": "rewriting", "topic": "code generation", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing code generation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating code generation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0191", "category": "rewriting", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing latency optimization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating latency optimization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0343", "category": "evaluation", "topic": "tokenization", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving tokenization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for tokenization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0164", "category": "summarization", "topic": "embedding search", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring embedding search. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot embedding search, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0279", "category": "glossary", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for classifier calibration, including one practical consideration."}, {"role": "assistant", "content": "Classifier calibration: checking whether predicted probabilities reflect observed outcome frequencies. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0035", "category": "glossary", "topic": "training validation", "messages": [{"role": "user", "content": "Write a short glossary entry for training validation, including one practical consideration."}, {"role": "assistant", "content": "Training validation: using held-out examples to assess model behavior during development. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0200", "category": "rewriting", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing synthetic datasets stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating synthetic datasets and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0423", "category": "rewriting", "topic": "tokenization", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing tokenization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating tokenization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0565", "category": "extraction", "topic": "quantization", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing quantization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"quantization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0586", "category": "evaluation", "topic": "batching", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving batching."}, {"role": "assistant", "content": "1. Record the test setup and baseline for batching.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0384", "category": "glossary", "topic": "embedding search", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for embedding search, including one practical consideration."}, {"role": "assistant", "content": "Embedding search: retrieving items using similarity between numerical vector representations. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0225", "category": "evaluation", "topic": "quantization", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving quantization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for quantization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0338", "category": "extraction", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing hallucination detection and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"hallucination detection\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0045", "category": "summarization", "topic": "quantization", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring quantization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot quantization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0249", "category": "definition", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a small research prototype: Explain prompt evaluation in one sentence for a beginner."}, {"role": "assistant", "content": "Prompt evaluation means testing how model outputs change in response to prompts."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0142", "category": "glossary", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for on-device AI, including one practical consideration."}, {"role": "assistant", "content": "On-device ai: executing AI models locally on a phone, laptop, or other device. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0494", "category": "definition", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For technical documentation: Explain fine-tuning in one sentence for a beginner."}, {"role": "assistant", "content": "Fine-tuning means continuing model training on a smaller task- or domain-specific dataset."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0190", "category": "rewriting", "topic": "code generation", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing code generation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating code generation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0098", "category": "extraction", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing hallucination detection and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"hallucination detection\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0021", "category": "glossary", "topic": "GPU inference", "messages": [{"role": "user", "content": "Write a short glossary entry for GPU inference, including one practical consideration."}, {"role": "assistant", "content": "Gpu inference: running a trained model on a graphics processor to produce predictions or outputs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0276", "category": "glossary", "topic": "context windows", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for context windows, including one practical consideration."}, {"role": "assistant", "content": "Context windows: the limit on how many input and generated tokens a model can consider in one request. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0573", "category": "extraction", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing retrieval augmentation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"retrieval augmentation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0490", "category": "definition", "topic": "code generation", "messages": [{"role": "user", "content": "For technical documentation: Explain code generation in one sentence for a beginner."}, {"role": "assistant", "content": "Code generation means producing programming code from natural-language or other input."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0251", "category": "definition", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a small research prototype: Explain latency optimization in one sentence for a beginner."}, {"role": "assistant", "content": "Latency optimization means reducing the delay between an input request and a generated response."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0095", "category": "extraction", "topic": "training validation", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing training validation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"training validation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0354", "category": "evaluation", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving fine-tuning."}, {"role": "assistant", "content": "1. Record the test setup and baseline for fine-tuning.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0445", "category": "extraction", "topic": "quantization", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing quantization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"quantization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0430", "category": "rewriting", "topic": "code generation", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing code generation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating code generation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0168", "category": "summarization", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring dataset curation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot dataset curation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0472", "category": "evaluation", "topic": "memory usage", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving memory usage."}, {"role": "assistant", "content": "1. Record the test setup and baseline for memory usage.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0547", "category": "rewriting", "topic": "model distillation", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing model distillation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model distillation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0553", "category": "rewriting", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing retrieval augmentation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating retrieval augmentation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0198", "category": "rewriting", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing hallucination detection stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating hallucination detection and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0406", "category": "summarization", "topic": "batching", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring batching. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot batching, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0270", "category": "glossary", "topic": "code generation", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for code generation, including one practical consideration."}, {"role": "assistant", "content": "Code generation: producing programming code from natural-language or other input. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0286", "category": "summarization", "topic": "batching", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring batching. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot batching, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0013", "category": "definition", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Explain retrieval augmentation in one sentence for a beginner."}, {"role": "assistant", "content": "Retrieval augmentation means supplementing generation with information retrieved from an external collection."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0426", "category": "rewriting", "topic": "batching", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing batching stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating batching and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0464", "category": "evaluation", "topic": "embedding search", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving embedding search."}, {"role": "assistant", "content": "1. Record the test setup and baseline for embedding search.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0542", "category": "rewriting", "topic": "on-device AI", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing on-device AI stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating on-device AI and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0377", "category": "definition", "topic": "model deployment", "messages": [{"role": "user", "content": "Answer concisely: Explain model deployment in one sentence for a beginner."}, {"role": "assistant", "content": "Model deployment means making a model available for use in an application or service."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0523", "category": "summarization", "topic": "tokenization", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring tokenization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot tokenization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0272", "category": "glossary", "topic": "memory usage", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for memory usage, including one practical consideration."}, {"role": "assistant", "content": "Memory usage: the amount of RAM or accelerator memory consumed during execution. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0303", "category": "rewriting", "topic": "tokenization", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing tokenization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating tokenization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0336", "category": "extraction", "topic": "context windows", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing context windows and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"context windows\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0396", "category": "glossary", "topic": "context windows", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for context windows, including one practical consideration."}, {"role": "assistant", "content": "Context windows: the limit on how many input and generated tokens a model can consider in one request. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0077", "category": "rewriting", "topic": "model deployment", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing model deployment stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model deployment and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0206", "category": "extraction", "topic": "batching", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing batching and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"batching\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0582", "category": "evaluation", "topic": "on-device AI", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving on-device AI."}, {"role": "assistant", "content": "1. Record the test setup and baseline for on-device AI.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0316", "category": "rewriting", "topic": "context windows", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing context windows stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating context windows and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0498", "category": "definition", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For technical documentation: Explain hallucination detection in one sentence for a beginner."}, {"role": "assistant", "content": "Hallucination detection means identifying generated claims that lack support or are incorrect."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0114", "category": "evaluation", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving fine-tuning."}, {"role": "assistant", "content": "1. Record the test setup and baseline for fine-tuning.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0009", "category": "definition", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Explain prompt evaluation in one sentence for a beginner."}, {"role": "assistant", "content": "Prompt evaluation means testing how model outputs change in response to prompts."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0219", "category": "extraction", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing classifier calibration and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"classifier calibration\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0089", "category": "extraction", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing prompt evaluation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"prompt evaluation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0297", "category": "summarization", "topic": "model deployment", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring model deployment. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model deployment, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0188", "category": "rewriting", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing dataset curation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating dataset curation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0166", "category": "summarization", "topic": "batching", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring batching. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot batching, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0589", "category": "evaluation", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving prompt evaluation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for prompt evaluation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0212", "category": "extraction", "topic": "memory usage", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing memory usage and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"memory usage\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0101", "category": "evaluation", "topic": "GPU inference", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving GPU inference."}, {"role": "assistant", "content": "1. Record the test setup and baseline for GPU inference.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0323", "category": "extraction", "topic": "tokenization", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing tokenization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"tokenization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0390", "category": "glossary", "topic": "code generation", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for code generation, including one practical consideration."}, {"role": "assistant", "content": "Code generation: producing programming code from natural-language or other input. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0388", "category": "glossary", "topic": "dataset curation", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for dataset curation, including one practical consideration."}, {"role": "assistant", "content": "Dataset curation: selecting, cleaning, and documenting examples used for training or evaluation. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0014", "category": "definition", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Explain fine-tuning in one sentence for a beginner."}, {"role": "assistant", "content": "Fine-tuning means continuing model training on a smaller task- or domain-specific dataset."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0274", "category": "glossary", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for fine-tuning, including one practical consideration."}, {"role": "assistant", "content": "Fine-tuning: continuing model training on a smaller task- or domain-specific dataset. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0465", "category": "evaluation", "topic": "quantization", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving quantization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for quantization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0367", "category": "definition", "topic": "model distillation", "messages": [{"role": "user", "content": "Answer concisely: Explain model distillation in one sentence for a beginner."}, {"role": "assistant", "content": "Model distillation means training a smaller student model using guidance from a stronger teacher."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0313", "category": "rewriting", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing retrieval augmentation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating retrieval augmentation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0028", "category": "glossary", "topic": "dataset curation", "messages": [{"role": "user", "content": "Write a short glossary entry for dataset curation, including one practical consideration."}, {"role": "assistant", "content": "Dataset curation: selecting, cleaning, and documenting examples used for training or evaluation. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0193", "category": "rewriting", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing retrieval augmentation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating retrieval augmentation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0292", "category": "summarization", "topic": "memory usage", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring memory usage. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot memory usage, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0242", "category": "definition", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a small research prototype: Explain on-device AI in one sentence for a beginner."}, {"role": "assistant", "content": "On-device ai means executing AI models locally on a phone, laptop, or other device."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0254", "category": "definition", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a small research prototype: Explain fine-tuning in one sentence for a beginner."}, {"role": "assistant", "content": "Fine-tuning means continuing model training on a smaller task- or domain-specific dataset."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0571", "category": "extraction", "topic": "latency optimization", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing latency optimization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"latency optimization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0447", "category": "extraction", "topic": "model distillation", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing model distillation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model distillation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0555", "category": "rewriting", "topic": "training validation", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing training validation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating training validation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0243", "category": "definition", "topic": "tokenization", "messages": [{"role": "user", "content": "For a small research prototype: Explain tokenization in one sentence for a beginner."}, {"role": "assistant", "content": "Tokenization means turning text into smaller units represented by numeric token IDs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0094", "category": "extraction", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing fine-tuning and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"fine-tuning\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0525", "category": "summarization", "topic": "quantization", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring quantization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot quantization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0120", "category": "evaluation", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving synthetic datasets."}, {"role": "assistant", "content": "1. Record the test setup and baseline for synthetic datasets.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0503", "category": "glossary", "topic": "tokenization", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for tokenization, including one practical consideration."}, {"role": "assistant", "content": "Tokenization: turning text into smaller units represented by numeric token IDs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0578", "category": "extraction", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing hallucination detection and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"hallucination detection\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0493", "category": "definition", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For technical documentation: Explain retrieval augmentation in one sentence for a beginner."}, {"role": "assistant", "content": "Retrieval augmentation means supplementing generation with information retrieved from an external collection."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0257", "category": "definition", "topic": "model deployment", "messages": [{"role": "user", "content": "For a small research prototype: Explain model deployment in one sentence for a beginner."}, {"role": "assistant", "content": "Model deployment means making a model available for use in an application or service."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0050", "category": "summarization", "topic": "code generation", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring code generation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot code generation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0580", "category": "extraction", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing synthetic datasets and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"synthetic datasets\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0355", "category": "evaluation", "topic": "training validation", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving training validation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for training validation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0252", "category": "definition", "topic": "memory usage", "messages": [{"role": "user", "content": "For a small research prototype: Explain memory usage in one sentence for a beginner."}, {"role": "assistant", "content": "Memory usage means the amount of RAM or accelerator memory consumed during execution."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0409", "category": "summarization", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring prompt evaluation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot prompt evaluation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0207", "category": "extraction", "topic": "model distillation", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing model distillation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model distillation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0156", "category": "glossary", "topic": "context windows", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for context windows, including one practical consideration."}, {"role": "assistant", "content": "Context windows: the limit on how many input and generated tokens a model can consider in one request. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0214", "category": "extraction", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing fine-tuning and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"fine-tuning\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0301", "category": "rewriting", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing GPU inference stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating GPU inference and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0486", "category": "definition", "topic": "batching", "messages": [{"role": "user", "content": "For technical documentation: Explain batching in one sentence for a beginner."}, {"role": "assistant", "content": "Batching means processing multiple requests together to improve hardware utilization."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0144", "category": "glossary", "topic": "embedding search", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for embedding search, including one practical consideration."}, {"role": "assistant", "content": "Embedding search: retrieving items using similarity between numerical vector representations. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0362", "category": "definition", "topic": "on-device AI", "messages": [{"role": "user", "content": "Answer concisely: Explain on-device AI in one sentence for a beginner."}, {"role": "assistant", "content": "On-device ai means executing AI models locally on a phone, laptop, or other device."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0482", "category": "definition", "topic": "on-device AI", "messages": [{"role": "user", "content": "For technical documentation: Explain on-device AI in one sentence for a beginner."}, {"role": "assistant", "content": "On-device ai means executing AI models locally on a phone, laptop, or other device."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0425", "category": "rewriting", "topic": "quantization", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing quantization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating quantization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0119", "category": "evaluation", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving classifier calibration."}, {"role": "assistant", "content": "1. Record the test setup and baseline for classifier calibration.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0314", "category": "rewriting", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing fine-tuning stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating fine-tuning and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0508", "category": "glossary", "topic": "dataset curation", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for dataset curation, including one practical consideration."}, {"role": "assistant", "content": "Dataset curation: selecting, cleaning, and documenting examples used for training or evaluation. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0133", "category": "definition", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a student project: Explain retrieval augmentation in one sentence for a beginner."}, {"role": "assistant", "content": "Retrieval augmentation means supplementing generation with information retrieved from an external collection."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0078", "category": "rewriting", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing hallucination detection stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating hallucination detection and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0146", "category": "glossary", "topic": "batching", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for batching, including one practical consideration."}, {"role": "assistant", "content": "Batching: processing multiple requests together to improve hardware utilization. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0158", "category": "glossary", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for hallucination detection, including one practical consideration."}, {"role": "assistant", "content": "Hallucination detection: identifying generated claims that lack support or are incorrect. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0205", "category": "extraction", "topic": "quantization", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing quantization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"quantization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0350", "category": "evaluation", "topic": "code generation", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving code generation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for code generation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0497", "category": "definition", "topic": "model deployment", "messages": [{"role": "user", "content": "For technical documentation: Explain model deployment in one sentence for a beginner."}, {"role": "assistant", "content": "Model deployment means making a model available for use in an application or service."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0535", "category": "summarization", "topic": "training validation", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring training validation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot training validation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0356", "category": "evaluation", "topic": "context windows", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving context windows."}, {"role": "assistant", "content": "1. Record the test setup and baseline for context windows.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0305", "category": "rewriting", "topic": "quantization", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing quantization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating quantization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0492", "category": "definition", "topic": "memory usage", "messages": [{"role": "user", "content": "For technical documentation: Explain memory usage in one sentence for a beginner."}, {"role": "assistant", "content": "Memory usage means the amount of RAM or accelerator memory consumed during execution."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0262", "category": "glossary", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for on-device AI, including one practical consideration."}, {"role": "assistant", "content": "On-device ai: executing AI models locally on a phone, laptop, or other device. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0022", "category": "glossary", "topic": "on-device AI", "messages": [{"role": "user", "content": "Write a short glossary entry for on-device AI, including one practical consideration."}, {"role": "assistant", "content": "On-device ai: executing AI models locally on a phone, laptop, or other device. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0037", "category": "glossary", "topic": "model deployment", "messages": [{"role": "user", "content": "Write a short glossary entry for model deployment, including one practical consideration."}, {"role": "assistant", "content": "Model deployment: making a model available for use in an application or service. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0402", "category": "summarization", "topic": "on-device AI", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring on-device AI. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot on-device AI, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0373", "category": "definition", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Answer concisely: Explain retrieval augmentation in one sentence for a beginner."}, {"role": "assistant", "content": "Retrieval augmentation means supplementing generation with information retrieved from an external collection."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0211", "category": "extraction", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing latency optimization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"latency optimization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0091", "category": "extraction", "topic": "latency optimization", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing latency optimization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"latency optimization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0333", "category": "extraction", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing retrieval augmentation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"retrieval augmentation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0266", "category": "glossary", "topic": "batching", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for batching, including one practical consideration."}, {"role": "assistant", "content": "Batching: processing multiple requests together to improve hardware utilization. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0263", "category": "glossary", "topic": "tokenization", "messages": [{"role": "user", "content": "For a small research prototype: Write a short glossary entry for tokenization, including one practical consideration."}, {"role": "assistant", "content": "Tokenization: turning text into smaller units represented by numeric token IDs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0457", "category": "extraction", "topic": "model deployment", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing model deployment and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model deployment\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0584", "category": "evaluation", "topic": "embedding search", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving embedding search."}, {"role": "assistant", "content": "1. Record the test setup and baseline for embedding search.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0079", "category": "rewriting", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing classifier calibration stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating classifier calibration and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0363", "category": "definition", "topic": "tokenization", "messages": [{"role": "user", "content": "Answer concisely: Explain tokenization in one sentence for a beginner."}, {"role": "assistant", "content": "Tokenization means turning text into smaller units represented by numeric token IDs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0302", "category": "rewriting", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing on-device AI stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating on-device AI and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0092", "category": "extraction", "topic": "memory usage", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing memory usage and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"memory usage\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0496", "category": "definition", "topic": "context windows", "messages": [{"role": "user", "content": "For technical documentation: Explain context windows in one sentence for a beginner."}, {"role": "assistant", "content": "Context windows means the limit on how many input and generated tokens a model can consider in one request."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0182", "category": "rewriting", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing on-device AI stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating on-device AI and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0520", "category": "glossary", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for synthetic datasets, including one practical consideration."}, {"role": "assistant", "content": "Synthetic datasets: collections of artificially generated examples rather than directly observed samples. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0545", "category": "rewriting", "topic": "quantization", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing quantization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating quantization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0458", "category": "extraction", "topic": "hallucination detection", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing hallucination detection and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"hallucination detection\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0566", "category": "extraction", "topic": "batching", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing batching and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"batching\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0202", "category": "extraction", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing on-device AI and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"on-device AI\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0515", "category": "glossary", "topic": "training validation", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for training validation, including one practical consideration."}, {"role": "assistant", "content": "Training validation: using held-out examples to assess model behavior during development. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0397", "category": "glossary", "topic": "model deployment", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for model deployment, including one practical consideration."}, {"role": "assistant", "content": "Model deployment: making a model available for use in an application or service. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0499", "category": "definition", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For technical documentation: Explain classifier calibration in one sentence for a beginner."}, {"role": "assistant", "content": "Classifier calibration means checking whether predicted probabilities reflect observed outcome frequencies."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0099", "category": "extraction", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing classifier calibration and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"classifier calibration\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0366", "category": "definition", "topic": "batching", "messages": [{"role": "user", "content": "Answer concisely: Explain batching in one sentence for a beginner."}, {"role": "assistant", "content": "Batching means processing multiple requests together to improve hardware utilization."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0379", "category": "definition", "topic": "classifier calibration", "messages": [{"role": "user", "content": "Answer concisely: Explain classifier calibration in one sentence for a beginner."}, {"role": "assistant", "content": "Classifier calibration means checking whether predicted probabilities reflect observed outcome frequencies."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0307", "category": "rewriting", "topic": "model distillation", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing model distillation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model distillation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0229", "category": "evaluation", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving prompt evaluation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for prompt evaluation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0422", "category": "rewriting", "topic": "on-device AI", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing on-device AI stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating on-device AI and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0581", "category": "evaluation", "topic": "GPU inference", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving GPU inference."}, {"role": "assistant", "content": "1. Record the test setup and baseline for GPU inference.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0370", "category": "definition", "topic": "code generation", "messages": [{"role": "user", "content": "Answer concisely: Explain code generation in one sentence for a beginner."}, {"role": "assistant", "content": "Code generation means producing programming code from natural-language or other input."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0347", "category": "evaluation", "topic": "model distillation", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving model distillation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model distillation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0197", "category": "rewriting", "topic": "model deployment", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing model deployment stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating model deployment and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0117", "category": "evaluation", "topic": "model deployment", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving model deployment."}, {"role": "assistant", "content": "1. Record the test setup and baseline for model deployment.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0063", "category": "rewriting", "topic": "tokenization", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing tokenization stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating tokenization and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0169", "category": "summarization", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For a student project: Summarize this note in a sentence: \"Our team is exploring prompt evaluation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot prompt evaluation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0253", "category": "definition", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "For a small research prototype: Explain retrieval augmentation in one sentence for a beginner."}, {"role": "assistant", "content": "Retrieval augmentation means supplementing generation with information retrieved from an external collection."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0544", "category": "rewriting", "topic": "embedding search", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing embedding search stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating embedding search and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0509", "category": "glossary", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for prompt evaluation, including one practical consideration."}, {"role": "assistant", "content": "Prompt evaluation: testing how model outputs change in response to prompts. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0594", "category": "evaluation", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving fine-tuning."}, {"role": "assistant", "content": "1. Record the test setup and baseline for fine-tuning.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0281", "category": "summarization", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring GPU inference. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot GPU inference, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0122", "category": "definition", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a student project: Explain on-device AI in one sentence for a beginner."}, {"role": "assistant", "content": "On-device ai means executing AI models locally on a phone, laptop, or other device."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0033", "category": "glossary", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Write a short glossary entry for retrieval augmentation, including one practical consideration."}, {"role": "assistant", "content": "Retrieval augmentation: supplementing generation with information retrieved from an external collection. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0124", "category": "definition", "topic": "embedding search", "messages": [{"role": "user", "content": "For a student project: Explain embedding search in one sentence for a beginner."}, {"role": "assistant", "content": "Embedding search means retrieving items using similarity between numerical vector representations."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0342", "category": "evaluation", "topic": "on-device AI", "messages": [{"role": "user", "content": "For a small research prototype: Give two responsible evaluation checks for a demonstration involving on-device AI."}, {"role": "assistant", "content": "1. Record the test setup and baseline for on-device AI.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0241", "category": "definition", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a small research prototype: Explain GPU inference in one sentence for a beginner."}, {"role": "assistant", "content": "Gpu inference means running a trained model on a graphics processor to produce predictions or outputs."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0093", "category": "extraction", "topic": "retrieval augmentation", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing retrieval augmentation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"retrieval augmentation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0116", "category": "evaluation", "topic": "context windows", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving context windows."}, {"role": "assistant", "content": "1. Record the test setup and baseline for context windows.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0016", "category": "definition", "topic": "context windows", "messages": [{"role": "user", "content": "Explain context windows in one sentence for a beginner."}, {"role": "assistant", "content": "Context windows means the limit on how many input and generated tokens a model can consider in one request."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0500", "category": "definition", "topic": "synthetic datasets", "messages": [{"role": "user", "content": "For technical documentation: Explain synthetic datasets in one sentence for a beginner."}, {"role": "assistant", "content": "Synthetic datasets means collections of artificially generated examples rather than directly observed samples."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0518", "category": "glossary", "topic": "hallucination detection", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for hallucination detection, including one practical consideration."}, {"role": "assistant", "content": "Hallucination detection: identifying generated claims that lack support or are incorrect. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0517", "category": "glossary", "topic": "model deployment", "messages": [{"role": "user", "content": "For technical documentation: Write a short glossary entry for model deployment, including one practical consideration."}, {"role": "assistant", "content": "Model deployment: making a model available for use in an application or service. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0554", "category": "rewriting", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing fine-tuning stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating fine-tuning and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0042", "category": "summarization", "topic": "on-device AI", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring on-device AI. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot on-device AI, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0375", "category": "definition", "topic": "training validation", "messages": [{"role": "user", "content": "Answer concisely: Explain training validation in one sentence for a beginner."}, {"role": "assistant", "content": "Training validation means using held-out examples to assess model behavior during development."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0524", "category": "summarization", "topic": "embedding search", "messages": [{"role": "user", "content": "For technical documentation: Summarize this note in a sentence: \"Our team is exploring embedding search. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot embedding search, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0385", "category": "glossary", "topic": "quantization", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for quantization, including one practical consideration."}, {"role": "assistant", "content": "Quantization: representing model weights or calculations using lower numerical precision. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0151", "category": "glossary", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for latency optimization, including one practical consideration."}, {"role": "assistant", "content": "Latency optimization: reducing the delay between an input request and a generated response. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0454", "category": "extraction", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing fine-tuning and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"fine-tuning\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0404", "category": "summarization", "topic": "embedding search", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring embedding search. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot embedding search, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0195", "category": "rewriting", "topic": "training validation", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing training validation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating training validation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0319", "category": "rewriting", "topic": "classifier calibration", "messages": [{"role": "user", "content": "For a small research prototype: Rewrite for clarity: \"We are doing classifier calibration stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating classifier calibration and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0474", "category": "evaluation", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving fine-tuning."}, {"role": "assistant", "content": "1. Record the test setup and baseline for fine-tuning.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0414", "category": "summarization", "topic": "fine-tuning", "messages": [{"role": "user", "content": "Answer concisely: Summarize this note in a sentence: \"Our team is exploring fine-tuning. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot fine-tuning, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0334", "category": "extraction", "topic": "fine-tuning", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing fine-tuning and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"fine-tuning\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0387", "category": "glossary", "topic": "model distillation", "messages": [{"role": "user", "content": "Answer concisely: Write a short glossary entry for model distillation, including one practical consideration."}, {"role": "assistant", "content": "Model distillation: training a smaller student model using guidance from a stronger teacher. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0595", "category": "evaluation", "topic": "training validation", "messages": [{"role": "user", "content": "For technical documentation: Give two responsible evaluation checks for a demonstration involving training validation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for training validation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0255", "category": "definition", "topic": "training validation", "messages": [{"role": "user", "content": "For a small research prototype: Explain training validation in one sentence for a beginner."}, {"role": "assistant", "content": "Training validation means using held-out examples to assess model behavior during development."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0217", "category": "extraction", "topic": "model deployment", "messages": [{"role": "user", "content": "For a student project: Extract the focus area and next step as JSON from: \"We are testing model deployment and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"model deployment\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0283", "category": "summarization", "topic": "tokenization", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring tokenization. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot tokenization, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0085", "category": "extraction", "topic": "quantization", "messages": [{"role": "user", "content": "Extract the focus area and next step as JSON from: \"We are testing quantization and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"quantization\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0461", "category": "evaluation", "topic": "GPU inference", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving GPU inference."}, {"role": "assistant", "content": "1. Record the test setup and baseline for GPU inference.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0421", "category": "rewriting", "topic": "GPU inference", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing GPU inference stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating GPU inference and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0471", "category": "evaluation", "topic": "latency optimization", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving latency optimization."}, {"role": "assistant", "content": "1. Record the test setup and baseline for latency optimization.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0372", "category": "definition", "topic": "memory usage", "messages": [{"role": "user", "content": "Answer concisely: Explain memory usage in one sentence for a beginner."}, {"role": "assistant", "content": "Memory usage means the amount of RAM or accelerator memory consumed during execution."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0448", "category": "extraction", "topic": "dataset curation", "messages": [{"role": "user", "content": "Answer concisely: Extract the focus area and next step as JSON from: \"We are testing dataset curation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"dataset curation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0226", "category": "evaluation", "topic": "batching", "messages": [{"role": "user", "content": "For a student project: Give two responsible evaluation checks for a demonstration involving batching."}, {"role": "assistant", "content": "1. Record the test setup and baseline for batching.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0248", "category": "definition", "topic": "dataset curation", "messages": [{"role": "user", "content": "For a small research prototype: Explain dataset curation in one sentence for a beginner."}, {"role": "assistant", "content": "Dataset curation means selecting, cleaning, and documenting examples used for training or evaluation."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0568", "category": "extraction", "topic": "dataset curation", "messages": [{"role": "user", "content": "For technical documentation: Extract the focus area and next step as JSON from: \"We are testing dataset curation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"dataset curation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0429", "category": "rewriting", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing prompt evaluation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating prompt evaluation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0546", "category": "rewriting", "topic": "batching", "messages": [{"role": "user", "content": "For technical documentation: Rewrite for clarity: \"We are doing batching stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating batching and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0424", "category": "rewriting", "topic": "embedding search", "messages": [{"role": "user", "content": "Answer concisely: Rewrite for clarity: \"We are doing embedding search stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating embedding search and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0131", "category": "definition", "topic": "latency optimization", "messages": [{"role": "user", "content": "For a student project: Explain latency optimization in one sentence for a beginner."}, {"role": "assistant", "content": "Latency optimization means reducing the delay between an input request and a generated response."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0112", "category": "evaluation", "topic": "memory usage", "messages": [{"role": "user", "content": "Give two responsible evaluation checks for a demonstration involving memory usage."}, {"role": "assistant", "content": "1. Record the test setup and baseline for memory usage.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0041", "category": "summarization", "topic": "GPU inference", "messages": [{"role": "user", "content": "Summarize this note in a sentence: \"Our team is exploring GPU inference. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot GPU inference, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0068", "category": "rewriting", "topic": "dataset curation", "messages": [{"role": "user", "content": "Rewrite for clarity: \"We are doing dataset curation stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating dataset curation and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0143", "category": "glossary", "topic": "tokenization", "messages": [{"role": "user", "content": "For a student project: Write a short glossary entry for tokenization, including one practical consideration."}, {"role": "assistant", "content": "Tokenization: turning text into smaller units represented by numeric token IDs. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0181", "category": "rewriting", "topic": "GPU inference", "messages": [{"role": "user", "content": "For a student project: Rewrite for clarity: \"We are doing GPU inference stuff and checking if it is good, maybe we try bigger later.\""}, {"role": "assistant", "content": "We are evaluating GPU inference and will consider expanding the experiment after reviewing the results."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0330", "category": "extraction", "topic": "code generation", "messages": [{"role": "user", "content": "For a small research prototype: Extract the focus area and next step as JSON from: \"We are testing code generation and will record latency and accuracy.\""}, {"role": "assistant", "content": "{\"focus_area\": \"code generation\", \"next_step\": \"record latency and accuracy\"}"}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0475", "category": "evaluation", "topic": "training validation", "messages": [{"role": "user", "content": "Answer concisely: Give two responsible evaluation checks for a demonstration involving training validation."}, {"role": "assistant", "content": "1. Record the test setup and baseline for training validation.\n2. Inspect failures and disclose limitations before drawing conclusions."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0015", "category": "definition", "topic": "training validation", "messages": [{"role": "user", "content": "Explain training validation in one sentence for a beginner."}, {"role": "assistant", "content": "Training validation means using held-out examples to assess model behavior during development."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0029", "category": "glossary", "topic": "prompt evaluation", "messages": [{"role": "user", "content": "Write a short glossary entry for prompt evaluation, including one practical consideration."}, {"role": "assistant", "content": "Prompt evaluation: testing how model outputs change in response to prompts. In practice, evaluate its benefits and trade-offs using measurements from the intended workload."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}
{"id": "compactlm-0287", "category": "summarization", "topic": "model distillation", "messages": [{"role": "user", "content": "For a small research prototype: Summarize this note in a sentence: \"Our team is exploring model distillation. We plan to test it on a small dataset first, measure the result, and document limitations before expanding.\""}, {"role": "assistant", "content": "The team will pilot model distillation, measure outcomes, and document limitations before scaling up."}], "synthetic": true, "source": "template-generated", "quality_note": "unreviewed demonstration sample"}