1 line
11 KiB
JSON
1 line
11 KiB
JSON
{"repo": "bairong-xdynamics/mistakenotebooklearning", "files": {"resources/agents/appworld/libs/appworld_agents/configs/simplified_react_code_agent/deepseek/deepseek-v3.2-terminus-exp-with-reasoning/dev.jsonnet": {"dc7de755522ad58864c62b74ab8e9959c01b7f23": {"t": 1785746374.92281, "n": 2308, "h": "a441d8be31c77f51", "c": "local experiment_prompts_path = std.extVar(\"APPWORLD_EXPERIMENT_PROMPTS_PATH\");\nlocal experiment_configs_path = std.extVar(\"APPWORLD_EXPERIMENT_CONFIGS_PATH\");\nlocal experiment_code_path = std.extVar(\"APPWORLD_EXPERIMENT_CODE_PATH\");\n{\n \"type\": \"simplified\",\n \"config\": {\n \"agent\": {\n \"type\": \"simplified_react_code_agent\",\n \"model_config\": {\n \"client_name\": \"openai\",\n \"api_type\": \"chat_completions\",\n \"base_url\": \"https://api.deepseek.com/v1\",\n \"api_key_env_name\": \"DEEPSEEK_API_KEY\",\n \"name\": \"deepseek-reasoner\",\n \"reasoning_effort\": \"high\",\n \"temperature\": 0.0,\n \"seed\": 100,\n \"drop_reasoning_content\": true,\n \"cost_per_token\": {\"input_cache_hit\": 2.8e-08, \"input_cache_miss\": 2.8e-07, \"input_cache_write\": 0.0, \"output\": 4.2e-07},\n \"retry_after_n_seconds\": 15,\n \"use_cache\": false,\n \"max_retries\": 100,\n },\n \"appworld_config\": {\n \"random_seed\": 100,\n \"raise_on_extra_parameters\": true,\n },\n \"logger_config\": {\n \"color\": true,\n \"verbose\": true,\n },\n \"usage_tracker_config\": {\n \"max_cost_overall\": 1000,\n \"max_cost_per_task\": 10,\n \"max_output_tokens_per_task\": 100000,\n },\n \"prompt_file_path\": experiment_prompts_path + \"/react_code_agent/instructions.txt\",\n \"ignore_multiple_calls\": true,\n \"max_prompt_length\": null,\n \"max_output_length\": null,\n \"max_steps\": 50,\n \"log_lm_calls\": true,\n \"skip_if_finished\": true,\n },\n \"dataset\": \"dev\",\n },\n \"metadata\": {\n \"model\": {\n \"file_name\": \"deepseek-v3.2-terminus-exp-with-reasoning\",\n \"humanized_name\": \"Deepseek V3.2 Terminus Exp With Reasoning\",\n \"precise_name\": \"deepseek-reasoner\",\n \"creator\": \"deepseek\",\n \"provider\": \"deepseek\",\n },\n \"agent\": {\n \"file_name\": \"simplified_react_code_agent\",\n \"humanized_name\": \"ReAct Code Agent\",\n },\n },\n}"}, "ef3506b9ee7dc1437a62ca619054c64e0b69bd5d": {"t": 1785763061.5824156, "n": 2308, "h": "a441d8be31c77f51", "c": "local experiment_prompts_path = std.extVar(\"APPWORLD_EXPERIMENT_PROMPTS_PATH\");\nlocal experiment_configs_path = std.extVar(\"APPWORLD_EXPERIMENT_CONFIGS_PATH\");\nlocal experiment_code_path = std.extVar(\"APPWORLD_EXPERIMENT_CODE_PATH\");\n{\n \"type\": \"simplified\",\n \"config\": {\n \"agent\": {\n \"type\": \"simplified_react_code_agent\",\n \"model_config\": {\n \"client_name\": \"openai\",\n \"api_type\": \"chat_completions\",\n \"base_url\": \"https://api.deepseek.com/v1\",\n \"api_key_env_name\": \"DEEPSEEK_API_KEY\",\n \"name\": \"deepseek-reasoner\",\n \"reasoning_effort\": \"high\",\n \"temperature\": 0.0,\n \"seed\": 100,\n \"drop_reasoning_content\": true,\n \"cost_per_token\": {\"input_cache_hit\": 2.8e-08, \"input_cache_miss\": 2.8e-07, \"input_cache_write\": 0.0, \"output\": 4.2e-07},\n \"retry_after_n_seconds\": 15,\n \"use_cache\": false,\n \"max_retries\": 100,\n },\n \"appworld_config\": {\n \"random_seed\": 100,\n \"raise_on_extra_parameters\": true,\n },\n \"logger_config\": {\n \"color\": true,\n \"verbose\": true,\n },\n \"usage_tracker_config\": {\n \"max_cost_overall\": 1000,\n \"max_cost_per_task\": 10,\n \"max_output_tokens_per_task\": 100000,\n },\n \"prompt_file_path\": experiment_prompts_path + \"/react_code_agent/instructions.txt\",\n \"ignore_multiple_calls\": true,\n \"max_prompt_length\": null,\n \"max_output_length\": null,\n \"max_steps\": 50,\n \"log_lm_calls\": true,\n \"skip_if_finished\": true,\n },\n \"dataset\": \"dev\",\n },\n \"metadata\": {\n \"model\": {\n \"file_name\": \"deepseek-v3.2-terminus-exp-with-reasoning\",\n \"humanized_name\": \"Deepseek V3.2 Terminus Exp With Reasoning\",\n \"precise_name\": \"deepseek-reasoner\",\n \"creator\": \"deepseek\",\n \"provider\": \"deepseek\",\n },\n \"agent\": {\n \"file_name\": \"simplified_react_code_agent\",\n \"humanized_name\": \"ReAct Code Agent\",\n },\n },\n}"}}, "resources/agents/appworld/libs/appworld_agents/configs/smolagents_tool_calling_agent/deepseek/deepseek-v3.2-terminus-exp-with-reasoning/dev.jsonnet": {"dc7de755522ad58864c62b74ab8e9959c01b7f23": {"t": 1785746493.4457424, "n": 2378, "h": "7173a19eb9c8a86e", "c": "local experiment_prompts_path = std.extVar(\"APPWORLD_EXPERIMENT_PROMPTS_PATH\");\nlocal experiment_configs_path = std.extVar(\"APPWORLD_EXPERIMENT_CONFIGS_PATH\");\nlocal experiment_code_path = std.extVar(\"APPWORLD_EXPERIMENT_CODE_PATH\");\n{\n \"type\": \"smolagents\",\n \"config\": {\n \"model\": {\n \"type\": \"openai\",\n \"base_url\": \"https://api.deepseek.com/v1\",\n \"api_key_env_name\": \"DEEPSEEK_API_KEY\",\n \"model_id\": \"deepseek-reasoner\",\n \"reasoning_effort\": \"high\",\n \"temperature\": 0.0,\n \"parallel_tool_calls\": true,\n \"seed\": 100,\n \"cost_per_token\": {\"input_cache_hit\": 2.8e-08, \"input_cache_miss\": 2.8e-07, \"input_cache_write\": 0.0, \"output\": 4.2e-07}, # NOTE: Not used, need to figure out how to use it.\n \"use_cache\": false,\n },\n \"api_predictor\": {\n \"mode\": \"predicted\",\n \"prompt_file_path\": experiment_prompts_path + \"/api_predictor.txt\",\n \"demo_task_ids\": [\n \"82e2fac_1\",\n \"29caf6f_1\",\n \"d0b1f43_1\"\n ],\n \"max_predicted_apis\": 16,\n },\n \"agent\": {\n \"type\": \"tool_calling\",\n \"prompt_templates\": experiment_prompts_path + \"/smolagents/tool_calling_instructions.yaml\",\n \"max_steps\": 50,\n \"max_seconds\": 500,\n \"max_cost_overall\": 1000,\n },\n \"task_completer\": {\n \"prompt_file_path\": experiment_prompts_path + \"/smolagents/task_completer_instructions.txt\",\n },\n \"appworld\": {\n \"random_seed\": 100,\n \"max_interactions\": 1000,\n \"max_api_calls_per_interaction\": 5000,\n \"raise_on_extra_parameters\": true,\n \"include_direct_functions\": true,\n \"direct_function_separator\": \"__\",\n },\n \"dataset\": \"dev\",\n },\n \"metadata\": {\n \"model\": {\n \"file_name\": \"deepseek-v3.2-terminus-exp-with-reasoning\",\n \"humanized_name\": \"Deepseek V3.2 Terminus Exp With Reasoning\",\n \"precise_name\": \"deepseek-reasoner\",\n \"creator\": \"deepseek\",\n \"provider\": \"deepseek\",\n },\n \"agent\": {\n \"file_name\": \"smolagents_tool_calling_agent\",\n \"humanized_name\": \"Smolagents ToolCallingAgent\",\n },\n },\n}"}, "ef3506b9ee7dc1437a62ca619054c64e0b69bd5d": {"t": 1785763061.3517368, "n": 2378, "h": "7173a19eb9c8a86e", "c": "local experiment_prompts_path = std.extVar(\"APPWORLD_EXPERIMENT_PROMPTS_PATH\");\nlocal experiment_configs_path = std.extVar(\"APPWORLD_EXPERIMENT_CONFIGS_PATH\");\nlocal experiment_code_path = std.extVar(\"APPWORLD_EXPERIMENT_CODE_PATH\");\n{\n \"type\": \"smolagents\",\n \"config\": {\n \"model\": {\n \"type\": \"openai\",\n \"base_url\": \"https://api.deepseek.com/v1\",\n \"api_key_env_name\": \"DEEPSEEK_API_KEY\",\n \"model_id\": \"deepseek-reasoner\",\n \"reasoning_effort\": \"high\",\n \"temperature\": 0.0,\n \"parallel_tool_calls\": true,\n \"seed\": 100,\n \"cost_per_token\": {\"input_cache_hit\": 2.8e-08, \"input_cache_miss\": 2.8e-07, \"input_cache_write\": 0.0, \"output\": 4.2e-07}, # NOTE: Not used, need to figure out how to use it.\n \"use_cache\": false,\n },\n \"api_predictor\": {\n \"mode\": \"predicted\",\n \"prompt_file_path\": experiment_prompts_path + \"/api_predictor.txt\",\n \"demo_task_ids\": [\n \"82e2fac_1\",\n \"29caf6f_1\",\n \"d0b1f43_1\"\n ],\n \"max_predicted_apis\": 16,\n },\n \"agent\": {\n \"type\": \"tool_calling\",\n \"prompt_templates\": experiment_prompts_path + \"/smolagents/tool_calling_instructions.yaml\",\n \"max_steps\": 50,\n \"max_seconds\": 500,\n \"max_cost_overall\": 1000,\n },\n \"task_completer\": {\n \"prompt_file_path\": experiment_prompts_path + \"/smolagents/task_completer_instructions.txt\",\n },\n \"appworld\": {\n \"random_seed\": 100,\n \"max_interactions\": 1000,\n \"max_api_calls_per_interaction\": 5000,\n \"raise_on_extra_parameters\": true,\n \"include_direct_functions\": true,\n \"direct_function_separator\": \"__\",\n },\n \"dataset\": \"dev\",\n },\n \"metadata\": {\n \"model\": {\n \"file_name\": \"deepseek-v3.2-terminus-exp-with-reasoning\",\n \"humanized_name\": \"Deepseek V3.2 Terminus Exp With Reasoning\",\n \"precise_name\": \"deepseek-reasoner\",\n \"creator\": \"deepseek\",\n \"provider\": \"deepseek\",\n },\n \"agent\": {\n \"file_name\": \"smolagents_tool_calling_agent\",\n \"humanized_name\": \"Smolagents ToolCallingAgent\",\n },\n },\n}"}}}} |