conversations listlengths 10 300 | agent stringclasses 1
value | model stringclasses 1
value | model_provider stringclasses 1
value | date stringlengths 27 27 | task stringlengths 10 25 | episode stringclasses 87
values | run_id stringclasses 15
values | trial_name stringlengths 19 34 | result stringclasses 3
values | trace_source stringclasses 9
values | teacher stringclasses 1
value | teacher_model stringclasses 1
value | task_group stringclasses 3
values | reward float64 0 1 โ | started_at stringlengths 27 27 | exception stringclasses 1
value | n_episodes int64 5 144 | n_turns_converted int64 4 142 | git_history_cmds bool 2
classes | saw_bug_patch bool 2
classes | sft_tokens int64 2.72k 193k | sft_asst_turns int64 5 143 | sft_truncated bool 2
classes | shortcut_flags listlengths 0 6 | shortcut_strong bool 2
classes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T11:53:33.334436Z | tezos-01039_copy3 | episode-76 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | tezos-01039_copy3__NvGzw7B | AgentTimeoutError | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T11:53:33.334436Z | AgentTimeoutError | 78 | 77 | false | false | 104,645 | 77 | true | [
"git_clone"
] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T04:38:18.796757Z | superuser-000324_copy0002 | episode-14 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-000324_copy0002__cm47ha3 | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T04:38:18.796757Z | null | 15 | 15 | false | false | 26,840 | 15 | false | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T05:28:03.635318Z | superuser-002052_copy0001 | episode-26 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-002052_copy0001__6J8Qqju | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T05:28:03.635318Z | null | 27 | 27 | false | false | 75,488 | 27 | true | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T05:18:58.876177Z | superuser-001721_copy0000 | episode-25 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-001721_copy0000__wc9WZj5 | AgentTimeoutError | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T05:18:58.876177Z | AgentTimeoutError | 27 | 26 | false | false | 75,195 | 26 | true | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T07:41:59.346618Z | superuser-007702_copy0001 | episode-21 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-007702_copy0001__RPKKs57 | AgentTimeoutError | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T07:41:59.346618Z | AgentTimeoutError | 23 | 22 | false | false | 55,169 | 22 | true | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T08:21:37.519164Z | superuser-008913_copy0002 | episode-7 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-008913_copy0002__b67svMU | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T08:21:37.519164Z | null | 8 | 8 | false | false | 11,765 | 8 | false | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T18:31:16.414012Z | tezos-0695_copy1 | episode-26 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | tezos-0695_copy1__Hv8PBBw | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T18:31:16.414012Z | null | 27 | 27 | false | false | 51,735 | 27 | true | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T17:24:17.876307Z | tezos-04635_copy8 | episode-42 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | tezos-04635_copy8__W6LBVhB | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T17:24:17.876307Z | null | 43 | 42 | false | false | 49,507 | 43 | true | [
"hist_cmd",
"git_clone"
] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T05:15:25.769244Z | superuser-001700_copy0001 | episode-6 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-001700_copy0001__6LHbfee | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T05:15:25.769244Z | null | 7 | 7 | false | false | 16,271 | 7 | false | [] | false |
[{"role":"user","content":"You are an AI assistant tasked with solving command-line tasks in a Linux(...TRUNCATED) | terminus-2 | hosted_vllm/t9-glm53-flash-openrouter | hosted_vllm | 2026-11-11T07:57:25.537252Z | superuser-008161_copy0001 | episode-8 | 607d9001-3f66-4924-b9a1-fbdfd6518147 | superuser-008161_copy0001__253bY2b | null | main | T9 | glm53-flash-openrouter | unverified | null | 2026-11-11T07:57:25.537252Z | null | 9 | 9 | false | false | 11,139 | 9 | false | [] | false |
OT-SFT-Data
Agent traces from several teachers on the OpenThoughts-Agent SFT task lists, collected for distillation experiments. The setup follows OpenThoughts-Agent: Terminus-2 (Harbor) with summarization on and no turn cap, 32,768-token context, one rollout per task, every trajectory kept (no success filter). Agent timeouts are 2x the task values, as in OpenThoughts-Agent. Teachers sample with their own default settings (temperature 1.0; Qwen teachers also top_p 0.95, top_k 20). Thinking is on, and earlier turns' reasoning is not sent back to the teacher. T9 (GLM 5.3 Flash) is served by OpenRouter with provider Z.AI (fp8) pinned and reasoning effort high, at the provider's default sampling settings, with each reply capped at 16,384 tokens (the other teachers ran on our own vLLM servers, where a reply was bounded only by the remaining context).
There are two kinds of config.
Trace sets (<teacher>-<size>): every exported row. Rows are in OpenThoughts-Agent's format (harbor traces export, last episode per trial): trace_source is main (the whole trajectory from the task prompt) or a
summarization sub-trace (summarization-<n>-summary / -questions / -answers). Assistant turns are
<think>reasoning</think> followed by Terminus-2's JSON response, rebuilt from the parsed step (Terminus-2 did not
keep raw responses), so whitespace and any stray text around the model's JSON are not preserved; turns the agent could
not parse are kept as written. Extra columns: teacher, teacher_model, task_group, reward (verifier reward;
none for SuperUser/Tezos tasks, whose verifiers do not test anything), exception, n_episodes, n_turns_converted,
and two broad flags for SWE-Smith answer leakage through git history (the task repos keep their "Bug Patch" commit):
git_history_cmds (the agent ran git show/diff/checkout against history) and saw_bug_patch (that commit appeared
in its terminal output). Our infrastructure details in terminal output (Docker hosts' IP addresses and hostnames,
host paths) are replaced with placeholders such as 203.0.113.x and /host/volume.
| Config | Teacher model | Tasks | Rows | Tasks by group | Path |
|---|---|---|---|---|---|
T1-3.16k |
qwen38-27b | 3136 | 4359 | issue 834, swesmith 805, unverified 1497 | T1-qwen38-27b/3.16k/train.parquet |
T2-3.16k |
qwen38-27b-abl | 3135 | 4845 | issue 833, swesmith 805, unverified 1497 | T2-qwen38-27b-abl/3.16k/train.parquet |
T3-3.16k |
glm47-flash | 3135 | 11663 | issue 834, swesmith 805, unverified 1496 | T3-glm47-flash/3.16k/train.parquet |
T4-3.16k |
glm47-flash-abl | 3136 | 11932 | issue 834, swesmith 805, unverified 1497 | T4-glm47-flash-abl/3.16k/train.parquet |
T9-3.16k |
glm53-flash-openrouter | 3136 | 8155 | issue 834, swesmith 805, unverified 1497 | T9-glm53-flash-openrouter/3.16k/train.parquet |
Training sets (<teacher>-<size>-sft-v<k>): the rows to train on, following OpenThoughts-Agent's recipe. One row
per task: main if it fits in 32,768 tokens, otherwise the latest summary/answers sub-trace that fits and has at
least 5 assistant turns, otherwise main; then rows with fewer than 5 assistant turns are dropped. Train with every
assistant turn's <think> text kept (LlamaFactory qwen3_nothink, as OpenThoughts-Agent's 8B students) and
cutoff_len: 32768, which truncates longer rows (OpenThoughts-Agent truncates; do not drop them, since dropping long
rows acts as a success filter). Token counts (sft_tokens, rendered as <|im_start|>role\ncontent<|im_end|> with the
Qwen3-8B tokenizer) and sft_truncated are included. OpenThoughts-Agent's recipe: full fine-tune, 7 epochs, lr 4e-5,
cosine, warmup 0.1, global batch 96.
Mixed-format training sets (<teacher>-<size>-sft-v1-mixed): the v0 rows (same rows, same shortcut flags), with
about half of the tasks (chosen by a hash of the task name, the same tasks for every teacher; main traces only)
rewritten into Qwen3's native tool-call format, so students also learn <tool_call> syntax (students trained on v0
write tool calls as bare JSON in tool-calling harnesses). Tool rows have a short system prompt and two tools,
bash_command(keystrokes, duration) and mark_task_complete(): thinking becomes reasoning_content, analysis/plan
the visible text, each command a tool call, the terminal output the tool response; replies that were not valid JSON
are dropped with the parse-error reply. Columns: format (tool_calls / terminus_json), messages (JSON,
OpenAI-style), tools (JSON or null), text (the row rendered for training: tool rows with Qwen3-8B's chat template
and tools, thinking kept after the last user message as at inference; JSON rows as v0), recomputed sft_tokens /
sft_asst_turns / sft_truncated, and sft_tool_calls; conversations stays in the JSON form. Train on the
assistant spans of text (<|im_start|>assistant .. <|im_end|>), or feed messages + tools to a trainer
that applies the chat template. Code: sft_tool_format.py, build_sft_mixed.py.
Shortcut flags (regex detectors on the agent's commands and the output that follows, counted only inside the first
32,768 tokens; shortcut_scan.py): shortcut_flags lists what the row shows: hist_cmd (reads git history),
saw_bug_diff (a history command printed a diff; on SWE-Smith the "Bug Patch" commit is the injected bug),
restore_hist (restores code from history), restore_tests (pulls the removed fail-to-pass tests out of history),
git_clone / pkg_download (fetches the upstream repo or a released package; on IssueTasks these contain the fix),
fetch_fix (fetches an upstream PR/commit/patch URL), grader_probe (looks at /logs/verifier, /tests, /solution,
reward.txt), harness_logs (reads the harness's own episode logs), edit_test (writes, edits or deletes files in a
tests/ directory or test config), skip_test (skip/xfail markers), complete_turn1 (marks the task complete before
seeing any output), false_complete (marks it complete with verifier reward 0). shortcut_strong is true for the
answer-revealing ones: a SWE-Smith history leak (saw_bug_diff, restore_hist, restore_tests), an IssueTasks
upstream fetch (git_clone, pkg_download), fetch_fix, grader_probe or skip_test. v0 keeps these rows; for a
filtered set use ds.filter(lambda r: not r["shortcut_strong"]). Detectors see commands only, so special-casing
tests inside written code is not flagged.
| Config | Teacher model | Rows | Truncated at 32K | shortcut_strong |
Rows by group | Path |
|---|---|---|---|---|---|---|
T1-3.16k-sft-v0 |
qwen38-27b | 2515 | 1094 | 721 | issue 684, swesmith 804, unverified 1027 | T1-qwen38-27b/3.16k-sft-v0/train.parquet |
T1-3.16k-sft-v1-mixed |
qwen38-27b | 2515 | 1051 | 721 | issue 684, swesmith 804, unverified 1027 | T1-qwen38-27b/3.16k-sft-v1-mixed/train.parquet |
T2-3.16k-sft-v0 |
qwen38-27b-abl | 2688 | 1393 | 802 | issue 646, swesmith 804, unverified 1238 | T2-qwen38-27b-abl/3.16k-sft-v0/train.parquet |
T2-3.16k-sft-v1-mixed |
qwen38-27b-abl | 2688 | 1348 | 802 | issue 646, swesmith 804, unverified 1238 | T2-qwen38-27b-abl/3.16k-sft-v1-mixed/train.parquet |
T3-3.16k-sft-v0 |
glm47-flash | 3053 | 286 | 178 | issue 823, swesmith 805, unverified 1425 | T3-glm47-flash/3.16k-sft-v0/train.parquet |
T3-3.16k-sft-v1-mixed |
glm47-flash | 3053 | 274 | 178 | issue 823, swesmith 805, unverified 1425 | T3-glm47-flash/3.16k-sft-v1-mixed/train.parquet |
T4-3.16k-sft-v0 |
glm47-flash-abl | 3044 | 253 | 141 | issue 822, swesmith 805, unverified 1417 | T4-glm47-flash-abl/3.16k-sft-v0/train.parquet |
T4-3.16k-sft-v1-mixed |
glm47-flash-abl | 3044 | 242 | 141 | issue 822, swesmith 805, unverified 1417 | T4-glm47-flash-abl/3.16k-sft-v1-mixed/train.parquet |
T9-3.16k-sft-v0 |
glm53-flash-openrouter | 2701 | 923 | 927 | issue 775, swesmith 803, unverified 1123 | T9-glm53-flash-openrouter/3.16k-sft-v0/train.parquet |
Load one: load_dataset("yixiaoh/OT-SFT-Data", "T1-3.16k-sft-v0"). Without a config name, load_dataset("yixiaoh/OT-SFT-Data") loads the
default config, T9-3.16k-sft-v0.
Task sources: OpenThoughts-Agent-SFT-3.16K and -10K task names, with task packages from DCAgent3/g1_diverse_tezos_top4_100k_sandboxes. IssueTasks come from SWE-bench issues; 282 of their source instances are in SWE-bench Verified, so exclude those before evaluating on SWE-bench Verified.
- Downloads last month
- 243