Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
episode_id: int64
task_name: string
pool: string
model: string
seed: int64
p1_steps: list<item: struct<step: int64, phase: int64, action: struct<action_type: string, target_service: str (... 2024 chars omitted)
  child 0, item: struct<step: int64, phase: int64, action: struct<action_type: string, target_service: string, parame (... 2012 chars omitted)
      child 0, step: int64
      child 1, phase: int64
      child 2, action: struct<action_type: string, target_service: string, parameters: struct<level: string, limit: int64,  (... 247 chars omitted)
          child 0, action_type: string
          child 1, target_service: string
          child 2, parameters: struct<level: string, limit: int64, root_cause: string, belief: struct<suspected_service: string, su (... 182 chars omitted)
              child 0, level: string
              child 1, limit: int64
              child 2, root_cause: string
              child 3, belief: struct<suspected_service: string, suspected_fault_class: string, service_confidence: double, fault_c (... 117 chars omitted)
                  child 0, suspected_service: string
                  child 1, suspected_fault_class: string
                  child 2, service_confidence: double
                  child 3, fault_confidence: double
                  child 4, evidence_gaps: list<item: string>
                      child 0, item: string
                  child 5, estimated_p2_cost: string
                  child 6, decision: string
                 
...
tring
num_p1_steps: int64
num_p2_steps: int64
cumulative_reward: double
score_breakdown: struct<final: double, p1_rca: double, p1_efficiency: double, patch_quality: double, no_change_detect (... 83 chars omitted)
  child 0, final: double
  child 1, p1_rca: double
  child 2, p1_efficiency: double
  child 3, patch_quality: double
  child 4, no_change_detection: double
  child 5, p2_efficiency: double
  child 6, r_cross: double
  child 7, null_context_p2_score: double
declared_patch: string
declared_no_change: bool
belief_history: list<item: struct<suspected_service: string, suspected_fault_class: string, service_confidence: doub (... 129 chars omitted)
  child 0, item: struct<suspected_service: string, suspected_fault_class: string, service_confidence: double, fault_c (... 117 chars omitted)
      child 0, suspected_service: string
      child 1, suspected_fault_class: string
      child 2, service_confidence: double
      child 3, fault_confidence: double
      child 4, evidence_gaps: list<item: string>
          child 0, item: string
      child 5, estimated_p2_cost: string
      child 6, decision: string
      child 7, reasoning: string
done: bool
worst_model: string
margin: double
rejected_reward: double
phase: int64
strategy: string
prompt: list<item: struct<role: string, content: string>>
  child 0, item: struct<role: string, content: string>
      child 0, role: string
      child 1, content: string
chosen: string
best_model: string
chosen_reward: double
rejected: string
to
{'prompt': List({'role': Value('string'), 'content': Value('string')}), 'chosen': Value('string'), 'rejected': Value('string'), 'chosen_reward': Value('float64'), 'rejected_reward': Value('float64'), 'margin': Value('float64'), 'task_name': Value('string'), 'pool': Value('string'), 'phase': Value('int64'), 'strategy': Value('string'), 'episode_id': Value('int64'), 'best_model': Value('string'), 'worst_model': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1779, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 295, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2281, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2227, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              episode_id: int64
              task_name: string
              pool: string
              model: string
              seed: int64
              p1_steps: list<item: struct<step: int64, phase: int64, action: struct<action_type: string, target_service: str (... 2024 chars omitted)
                child 0, item: struct<step: int64, phase: int64, action: struct<action_type: string, target_service: string, parame (... 2012 chars omitted)
                    child 0, step: int64
                    child 1, phase: int64
                    child 2, action: struct<action_type: string, target_service: string, parameters: struct<level: string, limit: int64,  (... 247 chars omitted)
                        child 0, action_type: string
                        child 1, target_service: string
                        child 2, parameters: struct<level: string, limit: int64, root_cause: string, belief: struct<suspected_service: string, su (... 182 chars omitted)
                            child 0, level: string
                            child 1, limit: int64
                            child 2, root_cause: string
                            child 3, belief: struct<suspected_service: string, suspected_fault_class: string, service_confidence: double, fault_c (... 117 chars omitted)
                                child 0, suspected_service: string
                                child 1, suspected_fault_class: string
                                child 2, service_confidence: double
                                child 3, fault_confidence: double
                                child 4, evidence_gaps: list<item: string>
                                    child 0, item: string
                                child 5, estimated_p2_cost: string
                                child 6, decision: string
                               
              ...
              tring
              num_p1_steps: int64
              num_p2_steps: int64
              cumulative_reward: double
              score_breakdown: struct<final: double, p1_rca: double, p1_efficiency: double, patch_quality: double, no_change_detect (... 83 chars omitted)
                child 0, final: double
                child 1, p1_rca: double
                child 2, p1_efficiency: double
                child 3, patch_quality: double
                child 4, no_change_detection: double
                child 5, p2_efficiency: double
                child 6, r_cross: double
                child 7, null_context_p2_score: double
              declared_patch: string
              declared_no_change: bool
              belief_history: list<item: struct<suspected_service: string, suspected_fault_class: string, service_confidence: doub (... 129 chars omitted)
                child 0, item: struct<suspected_service: string, suspected_fault_class: string, service_confidence: double, fault_c (... 117 chars omitted)
                    child 0, suspected_service: string
                    child 1, suspected_fault_class: string
                    child 2, service_confidence: double
                    child 3, fault_confidence: double
                    child 4, evidence_gaps: list<item: string>
                        child 0, item: string
                    child 5, estimated_p2_cost: string
                    child 6, decision: string
                    child 7, reasoning: string
              done: bool
              worst_model: string
              margin: double
              rejected_reward: double
              phase: int64
              strategy: string
              prompt: list<item: struct<role: string, content: string>>
                child 0, item: struct<role: string, content: string>
                    child 0, role: string
                    child 1, content: string
              chosen: string
              best_model: string
              chosen_reward: double
              rejected: string
              to
              {'prompt': List({'role': Value('string'), 'content': Value('string')}), 'chosen': Value('string'), 'rejected': Value('string'), 'chosen_reward': Value('float64'), 'rejected_reward': Value('float64'), 'margin': Value('float64'), 'task_name': Value('string'), 'pool': Value('string'), 'phase': Value('int64'), 'strategy': Value('string'), 'episode_id': Value('int64'), 'best_model': Value('string'), 'worst_model': Value('string')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
                  builder.download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 882, in download_and_prepare
                  self._download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 943, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1832, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

prompt
list
chosen
string
rejected
string
chosen_reward
float64
rejected_reward
float64
margin
float64
task_name
string
pool
string
phase
int64
strategy
string
episode_id
int64
best_model
string
worst_model
string
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "The root cause of the issue appears to be the recent deployment of orders v2.4.0, specifically the 'Refactor: extract pricing helper' change, which has led to degraded performance and increased memory usage in the api_gateway due to upstream issues wit...
0.13
-0.22
0.35
aliased_fault
A
1
within_episode
0
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "declare_no_change", "parameters": {"reason": "No relevant changes found in the code for the given commit."}}
{"action_type": "read_file", "parameters": {"path": "orders/orderservice.py"}}
0.08
-0.07
0.15
severity_inversion
B
2
within_episode
1
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
memory_leak
C
1
within_episode
2
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/orders/handlers/batch.py b/orders/handlers/batch.py\n--- a/orders/handlers/batch.py\n+++ b/orders/handlers/batch.py\n@@ -52,7 +52,7 @@ class BatchProcessor:\n self._notifier.send(order.id, \"submitted\")\n \n def cache_size(self) -> int:\...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
memory_leak
C
2
within_episode
2
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
memory_leak
A
1
within_episode
3
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
cascading_failure
C
1
within_episode
4
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/auth/config.py b/auth/config.py\n--- a/auth/config.py\n+++ b/auth/config.py\n@@ -8,6 +8,9 @@\n import os\n from dataclasses import dataclass\n \n+\n+_DEFAULT_DEV_SECRET = \"dev-only-do-not-use-in-prod-zZ3kf81P\"\n+\n \n @dataclass(frozen=True)\n clas...
{"action_type": "get_git_log", "parameters": {"path": ".", "n_commits": 15}}
0.08
0.03
0.05
cascading_failure
C
2
within_episode
4
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "The root cause of the issue appears to be a recent deploy of orders service version v3.0.0, which may have introduced a regression or bug affecting the checkout process."}}
0.13
-0.22
0.35
circuit_breaker_noop
C
1
within_episode
5
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "declare_no_change", "parameters": {"reason": "The commit d2b9c11 introduced the circuit breaker intentionally to prevent a known failure mode."}}
{"action_type": "read_file", "parameters": {"path": "orders/version.txt"}}
0.08
-0.07
0.15
circuit_breaker_noop
C
2
within_episode
5
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/orders/auth_client.py b/orders/auth_client.py\n--- a/orders/auth_client.py\n+++ b/orders/auth_client.py\n@@ -15,5 +15,6 @@ class AuthClient:\n def validate(self, token):\n- return self._call_with_retries(token, retries=2,\n- ...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
severity_inversion
B
2
within_episode
6
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Memory leak in payment service due to v1.9.0 deployment"}}
0.13
-0.22
0.35
confidence_inversion
C
1
within_episode
7
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/payment/threadpool.py b/payment/threadpool.py\n--- a/payment/threadpool.py\n+++ b/payment/threadpool.py\n@@ -8,4 +8,5 @@ class PoolWorker:\n def acquire(self):\n- self._lock_a.acquire()\n- self._lock_b.acquire()\n+ with self....
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
confidence_inversion
C
2
within_episode
7
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "The root cause of the issue appears to be a memory leak in the payment service, which was introduced in the recent deployment v3.0.0."}}
0.13
-0.22
0.35
circuit_breaker_noop
A
1
within_episode
8
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/payment/threadpool.py b/payment/threadpool.py\n--- a/payment/threadpool.py\n+++ b/payment/threadpool.py\n@@ -41,7 +41,7 @@ class PoolWorker:\n self._lock_a.acquire()\n self._lock_b.acquire()\n \n- def release(self) -> None:\n+ d...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
confidence_inversion
B
2
within_episode
9
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "auth service is down and causing degradation in api_gateway and orders services due to connection refusals."}}
0.13
-0.22
0.35
cascading_failure
A
1
within_episode
10
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
info_ordering
C
1
within_episode
11
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/shared/serializer.py b/shared/serializer.py\nindex 5b6c7d8..e7f8a9b 100644\n--- a/shared/serializer.py\n+++ b/shared/serializer.py\n@@ -23,7 +23,7 @@ from shared_serializer import encode # type: ignore\n \n class SchemaCodec:\n \"\"\"Per-service...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
info_ordering
C
2
within_episode
11
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "The root cause of the issue appears to be a Redis slow log in the auth service, likely due to a large number of session keys being scanned. This could be exacerbated by a recent deploy of the cache service."}}
0.13
-0.22
0.35
circuit_breaker_noop
C
1
within_episode
12
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "declare_no_change", "parameters": {"reason": "The added circuit breaker is intentional and correctly implemented. The fault is likely due to external factors or misinterpretation of the service behavior."}}
{"action_type": "list_dir", "parameters": {"path": "auth"}}
0.08
-0.07
0.15
circuit_breaker_noop
C
2
within_episode
12
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
info_ordering
A
1
within_episode
13
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Payment service is experiencing high error rates and latency, causing a cascade effect on dependent services like orders and api_gateway due to circuit breaker opening and connection refusals."}}
0.13
-0.22
0.35
distributed_deadlock
C
1
within_episode
14
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/payment/processor.py b/payment/processor.py\n--- a/payment/processor.py\n+++ b/payment/processor.py\n@@ -5,7 +5,7 @@\n \n from __future__ import annotations\n \n-import time\n+time,\n from typing import Optional\n \n from .gateway import StripeGatewa...
{"action_type": "get_git_log", "parameters": {"path": ".", "n_commits": 15}}
0.08
-0.07
0.15
distributed_deadlock
C
2
within_episode
14
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/orders/handlers/batch.py b/orders/handlers/batch.py\n--- a/orders/handlers/batch.py\n+++ b/orders/handlers/batch.py\n@@ -41,7 +41,7 @@\n self._cache[order.id] = order\n def flush(self) -> None:\n orders = list(self._cache.values()...
{"action_type": "get_git_log", "parameters": {"path": ".", "n_commits": 15}}
0.08
0.03
0.05
memory_leak
B
2
within_episode
15
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "--- a/payment/threadpool.py\n+++ b/payment/threadpool.py\n@@ -8,4 +8,5 @@ class PoolWorker:\n def acquire(self):\n- self._lock_a.acquire()\n- self._lock_b.acquire()\n+ with self._global_order:\n+ self._lock_a.acquire()\n+ ...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
confidence_inversion
B
2
within_episode
16
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Orders service experienced a memory leak starting from deploy v2.4.0, causing downstream services to degrade."}}
0.13
-0.22
0.35
aliased_fault
C
1
within_episode
17
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/queue/worker.py b/queue/worker.py\n--- a/queue/worker.py\n+++ b/queue/worker.py\n@@ -22,4 +22,5 @@ class CacheWriter:\n def flush(self, batch):\n- for k, v in batch.items():\n+ for k, v in list(batch.items()):\n self._ca...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
aliased_fault
C
2
within_episode
17
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "auth service is experiencing OOMKilled due to insufficient memory, leading to degraded status of dependent services api_gateway and orders."}}
0.13
-0.22
0.35
heldout_aliased_severity
D
1
within_episode
18
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "declare_no_change", "parameters": {"reason": "No relevant changes found in the provided commits and files."}}
{"action_type": "list_dir", "parameters": {"path": "auth"}}
0.08
-0.07
0.15
heldout_aliased_severity
D
2
within_episode
18
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/payment/threadpool.py b/payment/threadpool.py\n--- a/payment/threadpool.py\n+++ b/payment/threadpool.py\n@@ -8,4 +8,5 @@ class PoolWorker:\n def acquire(self):\n- self._lock_a.acquire()\n- self._lock_b.acquire()\n+ with self....
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
confidence_inversion
B
2
within_episode
19
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/auth/config.py b/auth/config.py\nindex 9a1b2c3..4d5e6f7 100644\n--- a/auth/config.py\n+++ b/auth/config.py\n@@ -14,7 +14,7 @@ AUTH_CONFIG = AuthConfig(\n token_ttl_seconds = 3600,\n issuer = \"auth.example.com\",\n )\n-\n+if __name...
{"action_type": "read_file", "parameters": {"path": "auth/auth_service.py"}}
0.08
-0.07
0.15
cascading_failure
B
2
within_episode
20
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Memory leak in payment service due to increased memory usage and high latency p99, accompanied by OOM exceptions and degraded service health."}}
0.13
-0.22
0.35
confidence_inversion
C
1
within_episode
21
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/payment/threadpool.py b/payment/threadpool.py\n--- a/payment/threadpool.py\n+++ b/payment/threadpool.py\n@@ -8,4 +8,5 @@ class PoolWorker:\n def acquire(self):\n- self._lock_a.acquire()\n- self._lock_b.acquire()\n+ with self....
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
confidence_inversion
C
2
within_episode
21
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "High message publish rate and queue depth issues in the payment service have cascaded to degrade the api_gateway."}}
0.13
-0.22
0.35
distributed_deadlock
A
1
within_episode
22
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
info_ordering
A
1
within_episode
23
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Payment service experienced a severe memory leak, leading to high memory usage and subsequent degradation of other services due to upstream failures."}}
0.13
-0.22
0.35
confidence_inversion
A
1
within_episode
24
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
cascading_failure
A
1
within_episode
25
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Payment service is experiencing high error rates and latency, leading to connection refusals from downstream services like orders."}}
0.13
-0.22
0.35
distributed_deadlock
C
1
within_episode
26
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/payment/processor.py b/payment/processor.py\n--- a/payment/processor.py\n+++ b/payment/processor.py\n@@ -23,7 +23,7 @@ class PaymentProcessor:\n self,\n gateway: StripeGateway,\n queue: QueueClient,\n- max_retries: in...
{"action_type": "get_git_log", "parameters": {"path": ".", "n_commits": 15}}
0.08
0.03
0.05
distributed_deadlock
C
2
within_episode
26
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Queue depth critical and publish flood detected for payment-service, causing degraded API Gateway response times and upstream service unavailability."}}
0.13
-0.22
0.35
distributed_deadlock
A
1
within_episode
27
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Payment service is experiencing a publish flood and queue depth critical, leading to degraded API Gateway performance."}}
0.13
-0.22
0.35
distributed_deadlock
A
1
within_episode
28
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "declare_root_cause", "parameters": {"root_cause": "Memory leak in orders service due to insufficient memory allocation and OOMKilled containers"}}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.38
-0.07
0.45
memory_leak
C
1
within_episode
29
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/orders/handlers/batch.py b/orders/handlers/batch.py\n--- a/orders/handlers/batch.py\n+++ b/orders/handlers/batch.py\n@@ -41,7 +41,7 @@\n self._cache[order.id] = order\n if len(self._cache) >= self._batch_size:\n self.flush...
{"action_type": "list_dir", "parameters": {"path": "."}}
0.08
0.03
0.05
memory_leak
C
2
within_episode
29
null
null
[ { "role": "system", "content": "You are an expert SRE handling a production incident in a microservices system.\n\n## Service Topology (downstream ← upstream)\napi_gateway ← auth, orders, cache\nauth ← database\norders ← database, payment, auth\npayment ← queue, database\ncache, database, queue ← (no depend...
{"action_type": "view_alerts"}
{"action_type": "query_logs", "target_service": "api_gateway", "parameters": {"level": "ERROR", "limit": 20}}
0.13
-0.07
0.2
cascading_failure
C
1
within_episode
30
null
null
[ { "role": "system", "content": "You are an expert SRE investigating a code-level fault in a sandboxed repository.\n\n## Phase 2 Action Space\nOutput EXACTLY ONE valid JSON action per turn. No markdown, no explanation.\n\nCode Exploration:\n{\"action_type\": \"list_dir\", \"parameters\": {\"path\": \".\"}}\n...
{"action_type": "propose_patch", "parameters": {"diff": "diff --git a/auth/config.py b/auth/config.py\n--- a/auth/config.py\n+++ b/auth/config.py\n@@ -8,6 +8,9 @@\n import os\n from dataclasses import dataclass\n \n+\n+_DEFAULT_DEV_SECRET = \"dev-only-do-not-use-in-prod-zZ3kf81P\"\n+\n \n @dataclass(frozen=True)\n clas...
{"action_type": "get_git_log", "parameters": {"path": ".", "n_commits": 15}}
0.08
0.03
0.05
cascading_failure
C
2
within_episode
30
null
null
End of preview.

No dataset card yet

Downloads last month
51

Space using sohambose98/sre-data 1