35ad3146a8
- Introduced new environments: Terminal Test Environment and SWE Environment, each with default configurations for testing and software engineering tasks. - Added TerminalBench 2.0 evaluation environment with comprehensive setup for agentic LLMs, including task execution and verification. - Enhanced ToolContext with methods for uploading and downloading files, ensuring binary-safe operations. - Updated documentation across environments to reflect new features and usage instructions. - Refactored existing environment configurations for consistency and clarity.
65 lines
1.9 KiB
YAML
65 lines
1.9 KiB
YAML
env:
|
|
group_size: 1
|
|
max_num_workers: -1
|
|
max_eval_workers: 16
|
|
max_num_workers_per_node: 8
|
|
steps_per_eval: 1
|
|
max_token_length: 32000
|
|
eval_handling: STOP_TRAIN
|
|
eval_limit_ratio: 0.5
|
|
inference_weight: 1.0
|
|
batch_size: -1
|
|
max_batches_offpolicy: 3
|
|
tokenizer_name: NousResearch/Hermes-3-Llama-3.1-8B
|
|
use_wandb: false
|
|
rollout_server_url: http://localhost:8000
|
|
total_steps: 1
|
|
wandb_name: terminal-bench-2
|
|
num_rollouts_to_keep: 32
|
|
num_rollouts_per_group_for_logging: 1
|
|
ensure_scores_are_not_same: false
|
|
data_path_to_save_groups: null
|
|
data_dir_to_save_evals: evals/terminal-bench-2
|
|
min_items_sent_before_logging: 2
|
|
include_messages: false
|
|
min_batch_allocation: null
|
|
worker_timeout: 600.0
|
|
thinking_mode: false
|
|
reasoning_effort: null
|
|
max_reasoning_tokens: null
|
|
custom_thinking_prompt: null
|
|
enabled_toolsets:
|
|
- terminal
|
|
- file
|
|
disabled_toolsets: null
|
|
distribution: null
|
|
max_agent_turns: 60
|
|
system_prompt: 'You are a skilled software engineer and system administrator with
|
|
access to a terminal and file tools. You are working inside a Linux container
|
|
environment. Complete the user''s task by using the available tools. Be methodical:
|
|
explore the environment first, plan your approach, then execute step by step.
|
|
Verify your work before finishing.'
|
|
agent_temperature: 1.0
|
|
terminal_backend: modal
|
|
dataset_name: NousResearch/terminal-bench-2
|
|
dataset_split: train
|
|
prompt_field: prompt
|
|
tool_call_parser: hermes
|
|
test_timeout: 180
|
|
force_build: false
|
|
task_filter: fix-git
|
|
skip_tasks: null
|
|
openai:
|
|
- timeout: 1200
|
|
num_max_requests_at_once: 512
|
|
num_requests_for_eval: 64
|
|
model_name: anthropic/claude-sonnet-4
|
|
rolling_buffer_length: 1000
|
|
server_type: openai
|
|
api_key: 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
|
|
base_url: https://openrouter.ai/api/v1
|
|
n_kwarg_is_ignored: false
|
|
health_check: false
|
|
slurm: false
|
|
testing: false
|