Skip to content

SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

Remove model config from anthropic_filemap_ml

Kabir Khandpur9bc034f62cda
diff --git a/config/anthropic_filemap_ml.yaml b/config/anthropic_filemap_ml.yaml
index e3cabeb9..e9a13e36 100644
--- a/config/anthropic_filemap_ml.yaml
+++ b/config/anthropic_filemap_ml.yaml
@@ -1,4 +1,6 @@
-# This template is heavily inspired by anthropic and openhands
+# This template is heavily inspired by anthropic and openhands. It is almost
+# identical to anthropic_filemap.yaml, but it removes python-specific language
+# and adds the ml_setup tool to support evaluation on the Multilingual dataset.
agent:
type: default
templates:
@@ -62,11 +64,3 @@ agent:
history_processors:
- type: cache_control
last_n_messages: 2
- model:
- name: claude-3-7-sonnet-20250219
- max_output_tokens: 64000
- per_instance_cost_limit: 2.50
- per_instance_call_limit: 150
- total_cost_limit: 500.0
- temperature: 0.0
- delay: 0.0