| Mode | Name | Size | |
|---|---|---|---|
| d--------- | anthropic_claude-opus-4.6 | 81 | logplain |
| d--------- | anthropic_claude-sonnet-4.6 | 81 | logplain |
| d--------- | deepseek_deepseek-v3.2 | 81 | logplain |
| -rw-r--r-- | experiment.log | 207507 | logplain |
| d--------- | moonshotai_kimi-k2.5 | 81 | logplain |
| d--------- | openai_gpt-5.3-codex | 91 | logplain |
| d--------- | qwen_qwen3-max-thinking | 81 | logplain |
| d--------- | qwen_qwen3.5-397b-a17b / child_protection / high | 43 | logplain |
| d--------- | qwen_qwen3.5-flash-02-23 / corporate_sabotage_v2 | 48 | logplain |
| -rw-r--r-- | results.csv | 145 | logplain |
| d--------- | z-ai_glm-4.7-flash | 123 | logplain |
 |
index : dissertation.git | |
| Undergraduate dissertation — Deceptive Alignment in Autonomous LLM Agents (UEL, 2026). Sandboxed agentic-LLM deception study: 960 runs, three-tier judging (regex + blackbox + glassbox). |
| summaryrefslogtreecommitdiff |
path: root/testing_different_models