Dataset Viewer
Auto-converted to Parquet Duplicate
conversations
listlengths
2
234
agent
stringclasses
1 value
model
stringclasses
1 value
model_provider
stringclasses
1 value
date
stringlengths
27
27
task
stringlengths
19
19
episode
stringclasses
57 values
run_id
stringclasses
1 value
trial_name
stringlengths
28
28
result
stringclasses
5 values
instruction
stringlengths
4
4.14k
verifier_output
stringlengths
330
102k
prompt_token_ids
listlengths
0
117
completion_token_ids
listlengths
0
117
logprobs
listlengths
0
117
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function that vali(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T14:17:53.945856Z
pymethods2test-1833
episode-6
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-1833__FryueTH
0.0
"Let me reconsider the test case for \"100000\". It has three consecutive \"00\" pairs (positions 1-(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,421,25419,38436,13513,440,3050,5426,25,198,16,13,15011,381,264,(...TRUNCATED)
[[-0.17496438324451447,-0.005939926952123642,-0.0005729932454414666,-0.000020742163542308845,0.0,-0.(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>I need to implement a function that calculates mini(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-17T23:46:40.757222Z
pymethods2test-4392
episode-1
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-4392__aSoHPQG
0.0
"The solution has been implemented in `/app/solution.py`. The function:\n\n1. **Reduces the input** (...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[40,1144,310,4072,264,709,421,45838,7776,10263,310,82391,7736,1608,8476,15019,11,1118,17191,23173,4(...TRUNCATED)
[[-1.1437149047851562,-0.0011108884355053306,-1.5497162166866474e-6,-0.07900877296924591,-0.03568491(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>I need to implement a function that simulates alien(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T22:35:11.151945Z
pymethods2test-0138
episode-5
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-0138__MniCtTz
0.0
"I need to implement the `calculate_blast_sequence` function. Let me analyze the requirements and cr(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[40,1144,310,4072,264,709,421,1586,22433,18995,6977,321,45838,10131,10263,13,6558,728,1118,1716,413(...TRUNCATED)
[[-1.721183180809021,-0.010452770628035069,-0.00009381314885104075,-0.5121992826461792,-0.0943890661(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function that chec(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T14:47:28.679455Z
pymethods2test-1708
episode-3
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-1708__hRPQDuw
0.0
"I'll implement the `is_passable_set` function using a Union-Find (Disjoint Set Union) data structur(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,421,11988,413,264,716,314,7552,7346,264,8173,3566,303,449,1962,(...TRUNCATED)
[[-0.05826963856816292,-0.023344917222857475,-0.0004515814362093806,-0.000020980565750505775,0.0,-6.(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>This is a classic optimization problem. I need to i(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T19:20:36.078259Z
pymethods2test-0716
episode-2
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-0716__Se7aKzq
0.0
"I'll implement a function to solve this optimization problem. This appears to be a variation of the(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[1919,369,264,11088,24460,3377,13,353,1144,310,4072,264,709,421,29044,4603,21803,33120,25152,2574,1(...TRUNCATED)
[[-1.0815894603729248,-0.1174549087882042,-0.04918678104877472,-0.6420888900756836,-0.08203154057264(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function that calc(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T02:12:35.092922Z
pymethods2test-3964
episode-2
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-3964__M8M7U9Q
0.0
"I'll implement the `max_nested_rubber_bands` function. Based on the problem description, this appea(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,421,45838,279,6966,1324,314,23283,21970,20254,303,264,4618,13,6(...TRUNCATED)
[[-0.8585060834884644,-0.004955509211868048,-0.02827048860490322,-0.00007128449942683801,-3.57627072(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function called `c(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T01:38:53.896618Z
pymethods2test-4067
episode-1
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-4067__PHQZcGq
1.0
"I've implemented the `calculate_decrement_percentage` function in `/app/solution.py`. The function:(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,2512,1510,34420,2187,13089,44507,63,440,3050,8242,13,6558,728,1(...TRUNCATED)
[[-0.0490415059030056,-0.004090277943760157,-0.0005968211917206645,-2.026550646405667e-6,0.0,-0.0004(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function `check_su(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T02:51:23.538044Z
pymethods2test-3840
episode-1
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-3840__toiUvhK
1.0
"Implementation complete. The `check_substring_presence` function:\n\n1. Swaps `text` and `pattern` (...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,1510,1960,5062,889,54509,63,303,36066,658,2600,3111,6971,27653,(...TRUNCATED)
[[-0.2664983868598938,-0.014481577090919018,-0.004349415190517902,-0.00018869027553591877,-8.3446315(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function that coun(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T19:33:12.080255Z
pymethods2test-0672
episode-1
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-0672__FDgb9qm
1.0
"I'll implement the function to count falling bumpers. Based on the examples, the logic is:\n- Count(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,421,14159,15218,26681,382,13,6558,728,3418,279,3377,25,271,16,1(...TRUNCATED)
[[-0.4392564296722412,-0.02074580267071724,-0.0011218428844586015,-0.000037788558984175324,0.0,-0.00(...TRUNCATED)
[{"content":"","role":"user"},{"content":"<think>The user wants me to implement a function that find(...TRUNCATED)
opencode
vllm/1393681232974600
vllm
2026-07-18T08:34:24.123854Z
pymethods2test-2907
episode-1
24ba0145-3d9f-4d4f-acee-93cf66253a0b
pymethods2test-2907__bhCenfe
1.0
"Done. The `find_unbalanced_tree` function:\n1. Returns `None` for empty input\n2. Counts occurrence(...TRUNCATED)
"Running tests...\n============================= test session starts ==============================\(...TRUNCATED)
[[248045,8678,198,2,13455,271,2523,599,2528,310,279,2614,5568,25,271,27,15449,29,198,4754,1267,763,3(...TRUNCATED)
[[760,1156,6587,728,310,4072,264,709,421,13325,279,3140,13418,29853,4757,12448,303,264,1103,13,6558,(...TRUNCATED)
[[-0.3355281949043274,-0.020590031519532204,-0.00035351174301467836,-0.000011205588634766173,0.0,-0.(...TRUNCATED)
End of preview. Expand in Data Studio

Agent trace dataset

Decoding the literal token IDs

The prompt_token_ids / completion_token_ids / logprobs columns are the verbatim tokens the serving engine emitted, stored PER AGENT STEP as a list-of-lists (one inner list per turn). To turn them back into text you MUST use the exact tokenizer the model was served with — a generic same-family tokenizer will decode word tokens to garbage.

Served model / tokenizer source: Qwen/Qwen3.5-122B-A10B-FP8

from transformers import AutoTokenizer
# Use the served model's own tokenizer (pull from the ref above; if it is a
# gs:// mirror, copy tokenizer.json/tokenizer_config.json/vocab.json/merges.txt
# locally first and point AutoTokenizer at that dir).
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-122B-A10B-FP8")
# token_ids are list-of-lists (one list per turn) — decode each turn:
text = [tok.decode(turn, skip_special_tokens=False) for turn in completion_token_ids]

Engine-reported served model name: 1393681232974600

See tokenizer_provenance.json for a machine-readable version.

Downloads last month
6