Publish Granite-4.2-8B aligned temperature-0.7 agentic results
Browse files[["scicoding-agent", 100, 33, 0], ["terminal-bench-2.1", 178, 32, 0]]; Replace both completed agentic exports; 2 explicitly user-authorized zero adjudications with original native null verdicts preserved. Reasoning exports unchanged.
- granite-4.2-8b/README.md +10 -20
- granite-4.2-8b/manifest.json +1 -1
- granite-4.2-8b/provenance.json +1 -1
- granite-4.2-8b/scicoding-agent/README.md +5 -0
- granite-4.2-8b/scicoding-agent/configs.json +0 -0
- granite-4.2-8b/scicoding-agent/judge_trajectories.jsonl +0 -0
- granite-4.2-8b/scicoding-agent/native_artifacts.tar.gz +2 -2
- granite-4.2-8b/scicoding-agent/provenance.json +1 -0
- granite-4.2-8b/scicoding-agent/summary.json +1 -1
- granite-4.2-8b/scicoding-agent/trajectories.jsonl +2 -2
- granite-4.2-8b/summary.json +1 -1
- granite-4.2-8b/terminal-bench-2.1/README.md +5 -0
- granite-4.2-8b/terminal-bench-2.1/configs.json +0 -0
- granite-4.2-8b/terminal-bench-2.1/judge_trajectories.jsonl +2 -2
- granite-4.2-8b/terminal-bench-2.1/native_artifacts.tar.gz +2 -2
- granite-4.2-8b/terminal-bench-2.1/provenance.json +1 -0
- granite-4.2-8b/terminal-bench-2.1/scoring_adjudications.json +1 -0
- granite-4.2-8b/terminal-bench-2.1/scoring_adjudications.tar.gz +3 -0
- granite-4.2-8b/terminal-bench-2.1/summary.json +1 -1
- granite-4.2-8b/terminal-bench-2.1/trajectories.jsonl +2 -2
granite-4.2-8b/README.md
CHANGED
|
@@ -1,26 +1,16 @@
|
|
| 1 |
-
#
|
| 2 |
|
| 3 |
-
|
| 4 |
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
-
|
| 8 |
|
| 9 |
-
|
| 10 |
|
| 11 |
-
|
| 12 |
-
|---|---|---:|---:|
|
| 13 |
-
| [arXiv Math](arxivmath/) | avg@20 | **26.65%** (549/2,060) | 0 unscored |
|
| 14 |
-
| [frontierscience-olympiad](frontierscience-olympiad/) | Accuracy | 32.30% | 0 |
|
| 15 |
-
| [frontierscience-research](frontierscience-research/) | Mean rubric | 20.97% | 1 |
|
| 16 |
-
| [scicoding-agent](scicoding-agent/) | avg@2 | 40.00% | 0 |
|
| 17 |
-
| [terminal-bench-2.1](terminal-bench-2.1/) | avg@2 | 17.98% | 0 |
|
| 18 |
-
| [hle-verified](hle-verified/) | Accuracy | 20.31% | 0 |
|
| 19 |
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
## Protocol and artifacts
|
| 23 |
-
|
| 24 |
-
Frozen model/hardware/sampling/agent budgets are in `configs.json`; source and completeness details are in `provenance.json`. These runs use their original frozen runtime, not an inferred current default. Agentic evaluations use avg@2, retain all 100 SciCoding or 178 TB attempts in the denominator, and count native verifier success even after an abnormal agent stop. Unscored results remain null, not fabricated native zeros. Research reports normalized rubric separately from threshold accuracy. HLE Gold and Revision are combined, with subset labels and original scores retained.
|
| 25 |
-
|
| 26 |
-
Each benchmark contains trajectories.jsonl, judge_trajectories.jsonl and summary.json. Prompt benchmarks also preserve event history; agentic benchmarks preserve native verifier logs and trajectory continuation files in native_artifacts.tar.gz. See manifest.json for file sizes and SHA-256 hashes.
|
|
|
|
| 1 |
+
# Granite-4.2-8B
|
| 2 |
|
| 3 |
+
## Current agentic evaluation (2026-11-16)
|
| 4 |
|
| 5 |
+
| Benchmark | Attempts | Passed | avg@2 | Unscored |
|
| 6 |
+
|---|---:|---:|---:|---:|
|
| 7 |
+
| scicoding-agent | 100 | 33 | 33.00% | 0 |
|
| 8 |
+
| terminal-bench-2.1 | 178 | 32 | 17.98% | 0 |
|
| 9 |
|
| 10 |
+
All attempts are finished and scored. Missing native verdicts were explicitly assigned zero by the user; their original null verdicts and errors are preserved alongside the adjudications. Every planned attempt stays in the denominator. Native verifier success counts regardless of agent stop reason. Exact unscored attempt IDs and native errors are retained in the exported judge records.
|
| 11 |
|
| 12 |
+
Terminus 2; temperature **0.7**, top-p **0.95**, top-k **20**; **16 H100 GPUs**, TP2, eight engines, two tasks per engine. Each benchmark contains full trajectories, verifier records, native artifacts, frozen configs and provenance. These exports replace the earlier agentic results; old exports remain recoverable in repository history.
|
| 13 |
|
| 14 |
+
## Unchanged reasoning and code evaluations
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
+
The current results and all files in [arxivmath](arxivmath/), [frontierscience-olympiad](frontierscience-olympiad/), [frontierscience-research](frontierscience-research/), [hle-verified](hle-verified/) and any previously published code-evaluation folders are preserved. Their own frozen configurations apply, not the agentic sampling settings above. See `summary.json` for the complete current results.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
granite-4.2-8b/manifest.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"README.md": {"size":
|
|
|
|
| 1 |
+
{"README.md": {"size": 1410, "sha256": "5e59644b58afec902a461685d20976733509d610115204dc079e47b27ec37696"}, "configs.json": {"size": 214472, "sha256": "6101bfca4421d2da51f5ee757e7d95e63930bfbc476a7be8d4851cc0fd5cf789"}, "provenance.json": {"size": 1558, "sha256": "27f5d6519fbb69cb84d4dcacfc00076adaf7c316165b825df705d5d2368bfc7b"}, "summary.json": {"size": 14853, "sha256": "14453d78b1770a1107c6a064a4de4ba87d9cd3cd629c97f96d9af7f29c33458f"}, "arxivmath/events.jsonl.gz": {"size": 135663416, "sha256": "5db63fe0b138ee438c5bf325209e19972cdfa698ff5bd3b241c9cfb631dbfaf2"}, "arxivmath/judge_trajectories.jsonl": {"size": 982015, "sha256": "df4c49ceb78b39e4a783a60b5046808f6c577f729319a0fe80bf50067ce80c45"}, "arxivmath/summary.json": {"size": 6093, "sha256": "132340f5a84f56f801b9fc897a6a00be9f382d2bb62ca2ccf8b1a01828c25813"}, "arxivmath/trajectories.jsonl": {"size": 260372976, "sha256": "b69675c85f99c76e963b1f41c27aa7527d734508593811486372a1b98649587b"}, "frontierscience-olympiad/events.jsonl.gz": {"size": 141512811, "sha256": "17ddd5e5526dc5714011792ae10dca35beac9c2ce19f4524071ec020a0886685"}, "frontierscience-olympiad/judge_trajectories.jsonl": {"size": 10037314, "sha256": "bcf78afc3344f69dd6b5c9ad4c7e7311248b91e3f16384a6d35e8d73771df681"}, "frontierscience-olympiad/summary.json": {"size": 1623, "sha256": "d54b3daee40ee4ad2c62ae1e2cbb4a66d5b861774e851b376bf085df9b575edd"}, "frontierscience-olympiad/trajectories.jsonl": {"size": 285406160, "sha256": "506fcb7581864460faf43731b3c4ffd85a696058b7de35c670c5a5c54caf6ff1"}, "frontierscience-research/events.jsonl.gz": {"size": 120759802, "sha256": "7ada6f7ef9192c57d5e2e5edb8c88be9d99dcdc13184433b3160a502dfdb33b8"}, "frontierscience-research/judge_trajectories.jsonl": {"size": 27120325, "sha256": "327b913f9d36ce1a95fba801266245ad4f81c03bdcfe9396e30ab30d97b2b5c2"}, "frontierscience-research/summary.json": {"size": 1814, "sha256": "33afcd5960122ae8e81499c821fc183fa8161009daf0f526e399ba2a9fbb3794"}, "frontierscience-research/trajectories.jsonl": {"size": 226759776, "sha256": "d32db79de4843c64b206ceb9bc273893b8762dba2f393c37129a53da8a773f9d"}, "hle-verified/events.jsonl.gz": {"size": 63041868, "sha256": "23ae9c6a27497bca6882d3e2039732f5f4f51e54cca03a0e282056f047658acf"}, "hle-verified/judge_trajectories.jsonl": {"size": 2366432, "sha256": "76c0d205c7fce84256bea34443f005531e9541aeecf49251b39a73d3f324081e"}, "hle-verified/summary.json": {"size": 3284, "sha256": "5030efa01ebd51009a22b4d0facc68fed01dba10d1ed24ceb57ef3f5c8d86b71"}, "hle-verified/trajectories.jsonl": {"size": 122573144, "sha256": "d57a25c2f714868dbd50befdd5698cb97e79e56fa5843f51f39ead4eb6c78be8"}, "arxivmath/README.md": {"size": 974, "sha256": "63220b73b3d31493b915d6605bed22cf018301a530507e741f80dc235f4c0d4e"}, "arxivmath/configs.json": {"size": 7616, "sha256": "c21c5d857ce71ad0be7898bb16266e55f285d8593c5400fa3ca87d5ac8a06f05"}, "arxivmath/provenance.json": {"size": 1287, "sha256": "989e33ada3bb53984033af89b1d2ce058cf98eae7be67550e26c9b0a89e98474"}, "scicoding-agent/README.md": {"size": 535, "sha256": "31f0dd3e998ef1219efd7ae216c4cc3719a4e4cb3101780b683ff30a5d1c9581"}, "scicoding-agent/configs.json": {"size": 215964, "sha256": "dad624f1aa7d3af383451f1a1c4bff2ec3ce309d5ae4ec42830d9b28e49e46ee"}, "scicoding-agent/judge_trajectories.jsonl": {"size": 343330, "sha256": "3ec8bcbe6c1272124ffff8d3320540a4ae9d39ec259369d5642da7feafa4e86a"}, "scicoding-agent/native_artifacts.tar.gz": {"size": 177712424, "sha256": "e04c8f81b0d9a738216c3b767db06d01c89f1975f2a5cf4856559dc45e957da2"}, "scicoding-agent/provenance.json": {"size": 373, "sha256": "049a7ce98a3770ef05384869dc4fe986906944b86666085a5cc1221a2291f5ad"}, "scicoding-agent/summary.json": {"size": 606, "sha256": "67c7ae2880baa1c8f05f93378dd5693162a45c7d3281a020a622ed65dbe1df26"}, "scicoding-agent/trajectories.jsonl": {"size": 49271423, "sha256": "15d3323de1eda09033c5ac9983062db4af080853d9212f4f02c7a13dc5659521"}, "terminal-bench-2.1/README.md": {"size": 538, "sha256": "05570042ad4d6dc58f862229a492d86e37e31948ca7096e9a30f9f548b949ca3"}, "terminal-bench-2.1/configs.json": {"size": 215964, "sha256": "dad624f1aa7d3af383451f1a1c4bff2ec3ce309d5ae4ec42830d9b28e49e46ee"}, "terminal-bench-2.1/judge_trajectories.jsonl": {"size": 5602321, "sha256": "9811baa2aabe0fbb155a45c0c7305ad885fa226a8f385e157a4eab0e065c6d5b"}, "terminal-bench-2.1/native_artifacts.tar.gz": {"size": 493642335, "sha256": "c60cfa61df83d25c0b960d207cfe7f4b79ebd02a42acf39482e8fb94ce3f22a7"}, "terminal-bench-2.1/provenance.json": {"size": 372, "sha256": "ffa1987a3905b895cf7bdf13397f6b9dea77af220106cb9ecd0eb0a67445d105"}, "terminal-bench-2.1/scoring_adjudications.json": {"size": 26817, "sha256": "6c9e67dfd8b8817473ba3cb9fa4a4aa6cbec031837a12ac6b2e3ebb0a8c44d8c"}, "terminal-bench-2.1/scoring_adjudications.tar.gz": {"size": 46746, "sha256": "7dd410307a0fb34f48cec7a0a7ce15bbcc34d390fb2207ee8c8873c29518f561"}, "terminal-bench-2.1/summary.json": {"size": 654, "sha256": "082d782fdd6bbcdbfe096a2d49d91dd8bd5c2d5f73cf3b0dada213fe2c41658e"}, "terminal-bench-2.1/trajectories.jsonl": {"size": 63305021, "sha256": "9804c92f45b541a09258f524573ef8e7e8c2f9d6550a6c5759030937db83d4a5"}}
|
granite-4.2-8b/provenance.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"source_run": "/workspace/data/sci-llm-training/runtime/model-eval-queue-20261105/granite42-8b/run", "queue_completed": {"time": 1794041760.927904, "result": "/workspace/data/sci-llm-training/runtime/model-eval-queue-20261105/granite42-8b/run/completed.json"}, "published_at": "2026-11-08", "regenerated_for_publication": false, "rejudged_for_publication": false, "unscored_records": {"arxivmath": [], "frontierscience-olympiad": [], "frontierscience-research": [{"key": "0b946e4d-8fc8-4f65-85dc-8fb7fd227a96::7", "task_id": "0b946e4d-8fc8-4f65-85dc-8fb7fd227a96", "attempt_id": null, "status": "score_error", "exception_info": null}], "scicoding-agent": [], "terminal-bench-2.1": [], "hle-verified": []}, "comparison_target": true, "runtime_policy": "Original frozen external-model runtime, explicitly approved; not relabeled as the newer pinned runtime.", "score_policy": "Native verifier results retained independently of abnormal agent completion. No new timeout-to-zero conversions.", "benchmark_overrides": {"arxivmath": {"path": "arxivmath/provenance.json", "aggregation": "avg@20", "superseded_revision": "8e48578e3102dc5486c26e05ebf8c7da74c2692e"}}}
|
|
|
|
| 1 |
+
{"source_run": "/workspace/data/sci-llm-training/runtime/model-eval-queue-20261105/granite42-8b/run", "queue_completed": {"time": 1794041760.927904, "result": "/workspace/data/sci-llm-training/runtime/model-eval-queue-20261105/granite42-8b/run/completed.json"}, "published_at": "2026-11-08", "regenerated_for_publication": false, "rejudged_for_publication": false, "unscored_records": {"arxivmath": [], "frontierscience-olympiad": [], "frontierscience-research": [{"key": "0b946e4d-8fc8-4f65-85dc-8fb7fd227a96::7", "task_id": "0b946e4d-8fc8-4f65-85dc-8fb7fd227a96", "attempt_id": null, "status": "score_error", "exception_info": null}], "scicoding-agent": [], "terminal-bench-2.1": [], "hle-verified": []}, "comparison_target": true, "runtime_policy": "Original frozen external-model runtime, explicitly approved; not relabeled as the newer pinned runtime.", "score_policy": "Native verifier results retained independently of abnormal agent completion. No new timeout-to-zero conversions.", "benchmark_overrides": {"arxivmath": {"path": "arxivmath/provenance.json", "aggregation": "avg@20", "superseded_revision": "8e48578e3102dc5486c26e05ebf8c7da74c2692e"}}, "latest_agentic_rerun": {"source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "published_date": "2026-11-16", "replaces_revision": "4e4dff1f8738756b67ef834f51e892443a6d3cd8", "benchmarks": ["scicoding-agent", "terminal-bench-2.1"], "configs": "See each agentic benchmark configs.json; prior provenance applies to unchanged benchmarks."}}
|
granite-4.2-8b/scicoding-agent/README.md
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# scicoding-agent: Granite-4.2-8B rerun
|
| 2 |
+
|
| 3 |
+
33/100 passed (33.00% avg@2); 0 unscored. All planned attempts remain in the denominator. Originally missing verdicts are explicitly user-adjudicated zero. Where applicable, original null verdicts and errors are preserved in scoring_adjudications.json and its backup archive.
|
| 4 |
+
|
| 5 |
+
Terminus 2; temperature 0.7, top-p 0.95, top-k 20; H100, TP2, eight engines, concurrency two per engine. Frozen plans and native artifacts are included. Native verifier success counts even after abnormal agent stops.
|
granite-4.2-8b/scicoding-agent/configs.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
granite-4.2-8b/scicoding-agent/judge_trajectories.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
granite-4.2-8b/scicoding-agent/native_artifacts.tar.gz
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e04c8f81b0d9a738216c3b767db06d01c89f1975f2a5cf4856559dc45e957da2
|
| 3 |
+
size 177712424
|
granite-4.2-8b/scicoding-agent/provenance.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "published_date": "2026-11-16", "replaces_revision": "4e4dff1f8738756b67ef834f51e892443a6d3cd8", "exporter_sha256": "3aec4e729f9b4b7d63fd9c183c07226a5e3350a7e5991a7e0c58738f3d4c0d15", "generation_or_judgment_rerun": false, "manual_zero_adjudications": false}
|
granite-4.2-8b/scicoding-agent/summary.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"tasks": 50, "attempts": 100, "attempts_per_task": 2, "metric": "avg@2", "passed":
|
|
|
|
| 1 |
+
{"tasks": 50, "attempts": 100, "attempts_per_task": 2, "metric": "avg@2", "passed": 33, "avg_at_2": 0.33, "unscored": 0, "agent_timeouts": 3, "reward_positive_with_exception": 0, "scoring": "native_verifier", "scoring_policy": "native-verifier-after-agent-stop-v1", "generation_finished": true, "grading_complete": true, "provisional": false, "source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "sampling": {"temperature": 0.7, "top_p": 0.95, "top_k": 20}, "hardware": "H100", "tasks_per_engine": 2, "adjudicated_zero": 0, "native_scored": 100}
|
granite-4.2-8b/scicoding-agent/trajectories.jsonl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:15d3323de1eda09033c5ac9983062db4af080853d9212f4f02c7a13dc5659521
|
| 3 |
+
size 49271423
|
granite-4.2-8b/summary.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"model": "granite42-8b", "benchmarks": {"arxivmath": {"benchmark": "arxivmath", "releases": [{"total_records": 340, "completed": 340, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.3, "accuracy": 0.3, "mean_latency_seconds": 1547.5274346591398, "finish_reason_counts": {"stop": 340}, "length_limited_response_count": 0, "observed_max_completion_tokens": 92448, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 17, "n_trajectories_scored": 340, "reported_accuracy": 0.3, "competition_metrics": {"arxiv/december": {"n_problems_scored": 17, "n_trajectories_scored": 340, "reported_accuracy": 0.3}}}, "benchmark": "matharena-arxivmath-december", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "c0efb2193ba5d71dbebb68ec7ffb2397a6bf0e24e4ae6f9599b6b033c88b7770", "scoring_hash": "24d031eb63c2c2608002961b52f16fdc1750f8bc97b29e03687c648b4197c87f", "selected": 340, "incomplete": 0, "generated_now": 340, "scored_now": 340, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 340, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T01:02:00.060547+00:00"}, {"total_records": 640, "completed": 640, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.2234375, "accuracy": 0.2234375, "mean_latency_seconds": 1472.8719204375902, "finish_reason_counts": {"stop": 640}, "length_limited_response_count": 0, "observed_max_completion_tokens": 90146, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 32, "n_trajectories_scored": 640, "reported_accuracy": 0.2234375, "competition_metrics": {"arxiv/february": {"n_problems_scored": 32, "n_trajectories_scored": 640, "reported_accuracy": 0.2234375}}}, "benchmark": "matharena-arxivmath-february", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "3d42ed80947d171b8b33ba172b5359039694683c86620ba5b4b9bba20e618add", "scoring_hash": "2ccec2e651751bd07ed5d3d34549ad91e6282fa85705d20ae0d3414100eb1e61", "selected": 640, "incomplete": 0, "generated_now": 640, "scored_now": 640, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 640, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T01:08:50.734824+00:00"}, {"total_records": 460, "completed": 460, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.3739130434782609, "accuracy": 0.3739130434782609, "mean_latency_seconds": 1711.2946264118652, "finish_reason_counts": {"stop": 460}, "length_limited_response_count": 0, "observed_max_completion_tokens": 79610, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 23, "n_trajectories_scored": 460, "reported_accuracy": 0.3739130434782609, "competition_metrics": {"arxiv/january": {"n_problems_scored": 23, "n_trajectories_scored": 460, "reported_accuracy": 0.3739130434782609}}}, "benchmark": "matharena-arxivmath-january", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "64946204331f5d3a4e7bb0e7e96bf9af915f073e79df5a3771f346b2aa62bab0", "scoring_hash": "802664108b12d31891570540a44593cfc6441ab51ec6cb2de3530fc58c8d6ebe", "selected": 460, "incomplete": 0, "generated_now": 460, "scored_now": 460, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 460, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T00:58:10.554391+00:00"}, {"total_records": 620, "completed": 620, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.2129032258064516, "accuracy": 0.2129032258064516, "mean_latency_seconds": 1146.9291893930979, "finish_reason_counts": {"stop": 620}, "length_limited_response_count": 0, "observed_max_completion_tokens": 85413, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 31, "n_trajectories_scored": 620, "reported_accuracy": 0.2129032258064516, "competition_metrics": {"arxiv/march": {"n_problems_scored": 31, "n_trajectories_scored": 620, "reported_accuracy": 0.2129032258064516}}}, "benchmark": "matharena-arxivmath-march", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "8023bb222faf2a5a66137726ebc97b0fd8721e7b4ef71f91f40187ca2ff67f5d", "scoring_hash": "86e5d7844f8275d29a3ec8d554310f9bfdac2ef43f374708395230485a7826ca", "selected": 620, "incomplete": 0, "generated_now": 620, "scored_now": 620, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 620, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T00:00:26.263342+00:00"}], "total_records": 2060, "completed": 2060, "accuracy": 0.26650485436893206, "correct": 549.0, "mean_score": 0.26650485436893206, "aggregation": "avg@20", "trials": 20, "questions": 103, "unscored": 0, "provisional": false, "publication_date": "2026-11-12", "source_run": "/workspace/data/sci-llm-training/runtime/arxiv-avg20-20261110/granite42-8b"}, "frontierscience-olympiad": {"total_records": 3000, "completed": 3000, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.323, "accuracy": 0.323, "mean_latency_seconds": 1131.3565981344755, "finish_reason_counts": {"length": 2, "stop": 2998}, "length_limited_response_count": 2, "observed_max_completion_tokens": 98304, "judge_usage": {"judged_record_count": 3000, "request_count": 3000, "retry_request_count": 0, "prompt_tokens": 2201361, "cached_prompt_tokens": 17225, "cache_write_prompt_tokens": 533882, "completion_tokens": 576346, "reasoning_tokens": 373815, "total_tokens": 2777707}, "judge_api_usage": {"request_count": 3000, "prompt_tokens": 2201361, "cached_prompt_tokens": 17225, "cache_write_prompt_tokens": 533882, "completion_tokens": 576346, "reasoning_tokens": 373815, "total_tokens": 2777707}, "benchmark_metrics": {"track": "olympiad", "n_problems_scored": 100, "n_trajectories_scored": 3000, "aggregation": "avg@30", "reported_accuracy": 0.32299999999999995, "mean_rubric_score": null, "success_threshold": 0.5, "community_profile": null}, "benchmark": "frontierscience-olympiad", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "89692ab508778aba3dd73c816a660486d69a024c097d66fa6c22110974ba167b", "scoring_hash": "0e899897c7cf45f80e15edc01a869181d6f588ca19d5dc3bcb855d8325f66209", "selected": 3000, "incomplete": 0, "generated_now": 3000, "scored_now": 3000, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 3000, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T21:34:06.491297+00:00"}, "frontierscience-research": {"total_records": 1800, "completed": 1799, "query_errors": 0, "score_errors": 1, "generated_unscored": 0, "mean_score": 0.20966328515842136, "accuracy": 0.03613118399110617, "mean_latency_seconds": 635.9828627359041, "finish_reason_counts": {"length": 4, "stop": 1795}, "length_limited_response_count": 4, "observed_max_completion_tokens": 98304, "judge_usage": {"judged_record_count": 1799, "request_count": 1819, "retry_request_count": 20, "prompt_tokens": 25213259, "cached_prompt_tokens": 249290, "cache_write_prompt_tokens": 24958512, "completion_tokens": 2909994, "reasoning_tokens": 1713028, "total_tokens": 28123253}, "judge_api_usage": {"request_count": 1818, "prompt_tokens": 25198049, "cached_prompt_tokens": 249290, "cache_write_prompt_tokens": 24943305, "completion_tokens": 2907853, "reasoning_tokens": 1711502, "total_tokens": 28105902}, "benchmark_metrics": {"track": "research", "n_problems_scored": 60, "n_trajectories_scored": 1799, "aggregation": "average_trials", "reported_accuracy": 0.03611111111111111, "mean_rubric_score": 0.2096148132183908, "success_threshold": 0.7, "community_profile": null}, "benchmark": "frontierscience-research", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "8ccd209deda790a0a7f7a50880fe20b494a62cd3b35c7be79b7bc0837501e82f", "scoring_hash": "e94267060b6ed01e455b1c25c69625896aaa44778fa60f6eddc9d577e40412b5", "selected": 1800, "incomplete": 1, "generated_now": 1800, "scored_now": 1798, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 1800, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T22:42:47.614049+00:00", "judgment_recovery": {"key": "a6176b58-f026-4fbc-9301-28190a36879c::4", "time": 1794011372.2380943, "generation_reused": true}}, "scicoding-agent": {"tasks": 50, "attempts": 100, "attempts_per_task": 2, "metric": "avg@2", "passed": 40, "avg_at_2": 0.4, "unscored": 0, "agent_timeouts": 1, "reward_positive_with_exception": 0, "scoring": "native_verifier", "scoring_policy": "native-verifier-after-agent-stop-v1"}, "terminal-bench-2.1": {"tasks": 89, "attempts": 178, "attempts_per_task": 2, "metric": "avg@2", "passed": 32, "avg_at_2": 0.1797752808988764, "unscored": 0, "agent_timeouts": 108, "reward_positive_with_exception": 5, "scoring": "native_verifier", "scoring_policy": "native-verifier-after-agent-stop-v1"}, "hle-verified": {"benchmark": "hle-verified", "total_records": 1600, "completed": 1600, "correct": 325.0, "accuracy": 0.203125, "mean_score": 0.203125, "aggregation": "question-weighted Gold + Revision, one trial per question", "subsets": {"gold": {"total_records": 575, "completed": 575, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.16695652173913045, "accuracy": 0.16695652173913045, "mean_latency_seconds": 551.8273080492417, "finish_reason_counts": {"stop": 575}, "length_limited_response_count": 0, "observed_max_completion_tokens": 97058, "judge_usage": {"judged_record_count": 575, "request_count": 575, "retry_request_count": 0, "prompt_tokens": 572682, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 271194, "completion_tokens": 66888, "reasoning_tokens": 32535, "total_tokens": 639570}, "judge_api_usage": {"request_count": 575, "prompt_tokens": 572682, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 271194, "completion_tokens": 66888, "reasoning_tokens": 32535, "total_tokens": 639570}, "benchmark_metrics": {"accuracy": 0.16695652173913045, "calibration_error": 66.56964347121034, "calibration_minimum_n": 100}, "benchmark": "hle-verified-gold", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "a223c02358042418016442a0cf7f2f4ce6a308edb49bf892d38761eb520d42cb", "scoring_hash": "b8c6522dcee1d6b19368b46b38acf19054303a135f9aa6be00d8341cc25ba783", "selected": 575, "incomplete": 0, "generated_now": 575, "scored_now": 575, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 575, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T17:10:04.985825+00:00"}, "revision": {"total_records": 1025, "completed": 1025, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.22341463414634147, "accuracy": 0.22341463414634147, "mean_latency_seconds": 755.2561215562611, "finish_reason_counts": {"stop": 1025}, "length_limited_response_count": 0, "observed_max_completion_tokens": 89232, "judge_usage": {"judged_record_count": 1025, "request_count": 1025, "retry_request_count": 0, "prompt_tokens": 1071852, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 572980, "completion_tokens": 142138, "reasoning_tokens": 77800, "total_tokens": 1213990}, "judge_api_usage": {"request_count": 1025, "prompt_tokens": 1071852, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 572980, "completion_tokens": 142138, "reasoning_tokens": 77800, "total_tokens": 1213990}, "benchmark_metrics": {"accuracy": 0.22341463414634147, "calibration_error": 63.39281131823441, "calibration_minimum_n": 100}, "benchmark": "hle-verified-revision", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "76686d6bfeafe3575d6cb2f9942a497d2dd8bc3c4bb4e2c266966fc740a8700b", "scoring_hash": "ca40769f8f5546ebd09487c03da114aef23995925a359df65bd29de710fde47c", "selected": 1025, "incomplete": 0, "generated_now": 1025, "scored_now": 1025, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 1025, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T17:59:17.289222+00:00"}}, "scoring": "LLM judge; original judgments retained"}}, "generation_finished": true, "fully_scored": false, "unscored": {"arxivmath": 0, "frontierscience-olympiad": 0, "frontierscience-research": 1, "scicoding-agent": 0, "terminal-bench-2.1": 0, "hle-verified": 0}, "arxiv_avg20_publication": {"date": "2026-11-12", "path": "arxivmath/", "correct": 549.0, "total": 2060}}
|
|
|
|
| 1 |
+
{"model": "granite42-8b", "benchmarks": {"arxivmath": {"benchmark": "arxivmath", "releases": [{"total_records": 340, "completed": 340, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.3, "accuracy": 0.3, "mean_latency_seconds": 1547.5274346591398, "finish_reason_counts": {"stop": 340}, "length_limited_response_count": 0, "observed_max_completion_tokens": 92448, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 17, "n_trajectories_scored": 340, "reported_accuracy": 0.3, "competition_metrics": {"arxiv/december": {"n_problems_scored": 17, "n_trajectories_scored": 340, "reported_accuracy": 0.3}}}, "benchmark": "matharena-arxivmath-december", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "c0efb2193ba5d71dbebb68ec7ffb2397a6bf0e24e4ae6f9599b6b033c88b7770", "scoring_hash": "24d031eb63c2c2608002961b52f16fdc1750f8bc97b29e03687c648b4197c87f", "selected": 340, "incomplete": 0, "generated_now": 340, "scored_now": 340, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 340, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T01:02:00.060547+00:00"}, {"total_records": 640, "completed": 640, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.2234375, "accuracy": 0.2234375, "mean_latency_seconds": 1472.8719204375902, "finish_reason_counts": {"stop": 640}, "length_limited_response_count": 0, "observed_max_completion_tokens": 90146, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 32, "n_trajectories_scored": 640, "reported_accuracy": 0.2234375, "competition_metrics": {"arxiv/february": {"n_problems_scored": 32, "n_trajectories_scored": 640, "reported_accuracy": 0.2234375}}}, "benchmark": "matharena-arxivmath-february", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "3d42ed80947d171b8b33ba172b5359039694683c86620ba5b4b9bba20e618add", "scoring_hash": "2ccec2e651751bd07ed5d3d34549ad91e6282fa85705d20ae0d3414100eb1e61", "selected": 640, "incomplete": 0, "generated_now": 640, "scored_now": 640, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 640, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T01:08:50.734824+00:00"}, {"total_records": 460, "completed": 460, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.3739130434782609, "accuracy": 0.3739130434782609, "mean_latency_seconds": 1711.2946264118652, "finish_reason_counts": {"stop": 460}, "length_limited_response_count": 0, "observed_max_completion_tokens": 79610, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 23, "n_trajectories_scored": 460, "reported_accuracy": 0.3739130434782609, "competition_metrics": {"arxiv/january": {"n_problems_scored": 23, "n_trajectories_scored": 460, "reported_accuracy": 0.3739130434782609}}}, "benchmark": "matharena-arxivmath-january", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "64946204331f5d3a4e7bb0e7e96bf9af915f073e79df5a3771f346b2aa62bab0", "scoring_hash": "802664108b12d31891570540a44593cfc6441ab51ec6cb2de3530fc58c8d6ebe", "selected": 460, "incomplete": 0, "generated_now": 460, "scored_now": 460, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 460, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T00:58:10.554391+00:00"}, {"total_records": 620, "completed": 620, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.2129032258064516, "accuracy": 0.2129032258064516, "mean_latency_seconds": 1146.9291893930979, "finish_reason_counts": {"stop": 620}, "length_limited_response_count": 0, "observed_max_completion_tokens": 85413, "judge_usage": {"judged_record_count": 0, "request_count": 0, "retry_request_count": 0, "prompt_tokens": 0, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 0, "completion_tokens": 0, "reasoning_tokens": 0, "total_tokens": 0}, "benchmark_metrics": {"competition": "arxivmath", "aggregation": "avg@20", "n_problems_scored": 31, "n_trajectories_scored": 620, "reported_accuracy": 0.2129032258064516, "competition_metrics": {"arxiv/march": {"n_problems_scored": 31, "n_trajectories_scored": 620, "reported_accuracy": 0.2129032258064516}}}, "benchmark": "matharena-arxivmath-march", "model": "granite42-8b-arxiv-avg20", "execution_mode": "non_agent", "generation_hash": "8023bb222faf2a5a66137726ebc97b0fd8721e7b4ef71f91f40187ca2ff67f5d", "scoring_hash": "86e5d7844f8275d29a3ec8d554310f9bfdac2ef43f374708395230485a7826ca", "selected": 620, "incomplete": 0, "generated_now": 620, "scored_now": 620, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 620, "generation_concurrency": 256, "judge_concurrency": 1, "updated_at": "2026-11-11T00:00:26.263342+00:00"}], "total_records": 2060, "completed": 2060, "accuracy": 0.26650485436893206, "correct": 549.0, "mean_score": 0.26650485436893206, "aggregation": "avg@20", "trials": 20, "questions": 103, "unscored": 0, "provisional": false, "publication_date": "2026-11-12", "source_run": "/workspace/data/sci-llm-training/runtime/arxiv-avg20-20261110/granite42-8b"}, "frontierscience-olympiad": {"total_records": 3000, "completed": 3000, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.323, "accuracy": 0.323, "mean_latency_seconds": 1131.3565981344755, "finish_reason_counts": {"length": 2, "stop": 2998}, "length_limited_response_count": 2, "observed_max_completion_tokens": 98304, "judge_usage": {"judged_record_count": 3000, "request_count": 3000, "retry_request_count": 0, "prompt_tokens": 2201361, "cached_prompt_tokens": 17225, "cache_write_prompt_tokens": 533882, "completion_tokens": 576346, "reasoning_tokens": 373815, "total_tokens": 2777707}, "judge_api_usage": {"request_count": 3000, "prompt_tokens": 2201361, "cached_prompt_tokens": 17225, "cache_write_prompt_tokens": 533882, "completion_tokens": 576346, "reasoning_tokens": 373815, "total_tokens": 2777707}, "benchmark_metrics": {"track": "olympiad", "n_problems_scored": 100, "n_trajectories_scored": 3000, "aggregation": "avg@30", "reported_accuracy": 0.32299999999999995, "mean_rubric_score": null, "success_threshold": 0.5, "community_profile": null}, "benchmark": "frontierscience-olympiad", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "89692ab508778aba3dd73c816a660486d69a024c097d66fa6c22110974ba167b", "scoring_hash": "0e899897c7cf45f80e15edc01a869181d6f588ca19d5dc3bcb855d8325f66209", "selected": 3000, "incomplete": 0, "generated_now": 3000, "scored_now": 3000, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 3000, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T21:34:06.491297+00:00"}, "frontierscience-research": {"total_records": 1800, "completed": 1799, "query_errors": 0, "score_errors": 1, "generated_unscored": 0, "mean_score": 0.20966328515842136, "accuracy": 0.03613118399110617, "mean_latency_seconds": 635.9828627359041, "finish_reason_counts": {"length": 4, "stop": 1795}, "length_limited_response_count": 4, "observed_max_completion_tokens": 98304, "judge_usage": {"judged_record_count": 1799, "request_count": 1819, "retry_request_count": 20, "prompt_tokens": 25213259, "cached_prompt_tokens": 249290, "cache_write_prompt_tokens": 24958512, "completion_tokens": 2909994, "reasoning_tokens": 1713028, "total_tokens": 28123253}, "judge_api_usage": {"request_count": 1818, "prompt_tokens": 25198049, "cached_prompt_tokens": 249290, "cache_write_prompt_tokens": 24943305, "completion_tokens": 2907853, "reasoning_tokens": 1711502, "total_tokens": 28105902}, "benchmark_metrics": {"track": "research", "n_problems_scored": 60, "n_trajectories_scored": 1799, "aggregation": "average_trials", "reported_accuracy": 0.03611111111111111, "mean_rubric_score": 0.2096148132183908, "success_threshold": 0.7, "community_profile": null}, "benchmark": "frontierscience-research", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "8ccd209deda790a0a7f7a50880fe20b494a62cd3b35c7be79b7bc0837501e82f", "scoring_hash": "e94267060b6ed01e455b1c25c69625896aaa44778fa60f6eddc9d577e40412b5", "selected": 1800, "incomplete": 1, "generated_now": 1800, "scored_now": 1798, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 1800, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T22:42:47.614049+00:00", "judgment_recovery": {"key": "a6176b58-f026-4fbc-9301-28190a36879c::4", "time": 1794011372.2380943, "generation_reused": true}}, "scicoding-agent": {"tasks": 50, "attempts": 100, "attempts_per_task": 2, "metric": "avg@2", "passed": 33, "avg_at_2": 0.33, "unscored": 0, "agent_timeouts": 3, "reward_positive_with_exception": 0, "scoring": "native_verifier", "scoring_policy": "native-verifier-after-agent-stop-v1", "generation_finished": true, "grading_complete": true, "provisional": false, "source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "sampling": {"temperature": 0.7, "top_p": 0.95, "top_k": 20}, "hardware": "H100", "tasks_per_engine": 2, "adjudicated_zero": 0, "native_scored": 100}, "terminal-bench-2.1": {"tasks": 89, "attempts": 178, "attempts_per_task": 2, "metric": "avg@2", "passed": 32, "avg_at_2": 0.1797752808988764, "unscored": 0, "agent_timeouts": 103, "reward_positive_with_exception": 3, "scoring": "native_verifier_with_explicit_user_adjudication", "scoring_policy": "native-verifier-after-agent-stop-v1", "generation_finished": true, "grading_complete": true, "provisional": false, "source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "sampling": {"temperature": 0.7, "top_p": 0.95, "top_k": 20}, "hardware": "H100", "tasks_per_engine": 2, "adjudicated_zero": 2, "native_scored": 176}, "hle-verified": {"benchmark": "hle-verified", "total_records": 1600, "completed": 1600, "correct": 325.0, "accuracy": 0.203125, "mean_score": 0.203125, "aggregation": "question-weighted Gold + Revision, one trial per question", "subsets": {"gold": {"total_records": 575, "completed": 575, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.16695652173913045, "accuracy": 0.16695652173913045, "mean_latency_seconds": 551.8273080492417, "finish_reason_counts": {"stop": 575}, "length_limited_response_count": 0, "observed_max_completion_tokens": 97058, "judge_usage": {"judged_record_count": 575, "request_count": 575, "retry_request_count": 0, "prompt_tokens": 572682, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 271194, "completion_tokens": 66888, "reasoning_tokens": 32535, "total_tokens": 639570}, "judge_api_usage": {"request_count": 575, "prompt_tokens": 572682, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 271194, "completion_tokens": 66888, "reasoning_tokens": 32535, "total_tokens": 639570}, "benchmark_metrics": {"accuracy": 0.16695652173913045, "calibration_error": 66.56964347121034, "calibration_minimum_n": 100}, "benchmark": "hle-verified-gold", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "a223c02358042418016442a0cf7f2f4ce6a308edb49bf892d38761eb520d42cb", "scoring_hash": "b8c6522dcee1d6b19368b46b38acf19054303a135f9aa6be00d8341cc25ba783", "selected": 575, "incomplete": 0, "generated_now": 575, "scored_now": 575, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 575, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T17:10:04.985825+00:00"}, "revision": {"total_records": 1025, "completed": 1025, "query_errors": 0, "score_errors": 0, "generated_unscored": 0, "mean_score": 0.22341463414634147, "accuracy": 0.22341463414634147, "mean_latency_seconds": 755.2561215562611, "finish_reason_counts": {"stop": 1025}, "length_limited_response_count": 0, "observed_max_completion_tokens": 89232, "judge_usage": {"judged_record_count": 1025, "request_count": 1025, "retry_request_count": 0, "prompt_tokens": 1071852, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 572980, "completion_tokens": 142138, "reasoning_tokens": 77800, "total_tokens": 1213990}, "judge_api_usage": {"request_count": 1025, "prompt_tokens": 1071852, "cached_prompt_tokens": 0, "cache_write_prompt_tokens": 572980, "completion_tokens": 142138, "reasoning_tokens": 77800, "total_tokens": 1213990}, "benchmark_metrics": {"accuracy": 0.22341463414634147, "calibration_error": 63.39281131823441, "calibration_minimum_n": 100}, "benchmark": "hle-verified-revision", "model": "granite42-8b", "execution_mode": "non_agent", "generation_hash": "76686d6bfeafe3575d6cb2f9942a497d2dd8bc3c4bb4e2c266966fc740a8700b", "scoring_hash": "ca40769f8f5546ebd09487c03da114aef23995925a359df65bd29de710fde47c", "selected": 1025, "incomplete": 0, "generated_now": 1025, "scored_now": 1025, "cached_completed": 0, "stage": "all", "worker_count": 1, "worker_index": 0, "worker_owned_predictions": 1025, "generation_concurrency": 256, "judge_concurrency": 256, "updated_at": "2026-11-06T17:59:17.289222+00:00"}}, "scoring": "LLM judge; original judgments retained"}}, "generation_finished": true, "fully_scored": false, "unscored": {"arxivmath": 0, "frontierscience-olympiad": 0, "frontierscience-research": 1, "scicoding-agent": 0, "terminal-bench-2.1": 0, "hle-verified": 0}, "arxiv_avg20_publication": {"date": "2026-11-12", "path": "arxivmath/", "correct": 549.0, "total": 2060}, "grading_complete": true, "provisional": false, "latest_agentic_rerun": {"published_at": "2026-11-16", "source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "benchmarks": ["scicoding-agent", "terminal-bench-2.1"], "unscored": 0}}
|
granite-4.2-8b/terminal-bench-2.1/README.md
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# terminal-bench-2.1: Granite-4.2-8B rerun
|
| 2 |
+
|
| 3 |
+
32/178 passed (17.98% avg@2); 0 unscored. All planned attempts remain in the denominator. Originally missing verdicts are explicitly user-adjudicated zero. Where applicable, original null verdicts and errors are preserved in scoring_adjudications.json and its backup archive.
|
| 4 |
+
|
| 5 |
+
Terminus 2; temperature 0.7, top-p 0.95, top-k 20; H100, TP2, eight engines, concurrency two per engine. Frozen plans and native artifacts are included. Native verifier success counts even after abnormal agent stops.
|
granite-4.2-8b/terminal-bench-2.1/configs.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
granite-4.2-8b/terminal-bench-2.1/judge_trajectories.jsonl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9811baa2aabe0fbb155a45c0c7305ad885fa226a8f385e157a4eab0e065c6d5b
|
| 3 |
+
size 5602321
|
granite-4.2-8b/terminal-bench-2.1/native_artifacts.tar.gz
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c60cfa61df83d25c0b960d207cfe7f4b79ebd02a42acf39482e8fb94ce3f22a7
|
| 3 |
+
size 493642335
|
granite-4.2-8b/terminal-bench-2.1/provenance.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "published_date": "2026-11-16", "replaces_revision": "4e4dff1f8738756b67ef834f51e892443a6d3cd8", "exporter_sha256": "3aec4e729f9b4b7d63fd9c183c07226a5e3350a7e5991a7e0c58738f3d4c0d15", "generation_or_judgment_rerun": false, "manual_zero_adjudications": true}
|
granite-4.2-8b/terminal-bench-2.1/scoring_adjudications.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"headless-terminal--attempt-001": {"method": "explicit_user_authorized_zero", "authorization": "[REDACTED_CREDENTIAL]", "timestamp": "2026-11-16T16:48:15.753731+00:00", "reward": 0.0, "native_reward": null, "original_verifier_result": null, "original_exception_info": {"exception_type": "AgentTimeoutError", "exception_message": "Agent execution timed out after 900.0 seconds", "exception_traceback": "Traceback (most recent call last):\n File \"/usr/local/lib/python3.12/asyncio/tasks.py\", line 520, in wait_for\n return await fut\n ^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1743, in run\n await self._run_agent_loop(\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1410, in _run_agent_loop\n ) = await self._handle_llm_interaction(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1288, in _handle_llm_interaction\n llm_response = await self._query_llm(\n ^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 111, in query\n response = await retry(checked, *args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 126, in __call__\n do = await self.iter(retry_state=retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 171, in iter\n result = await action(retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/_utils.py\", line 130, in inner\n return call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/__init__.py\", line 462, in <lambda>\n self._add_action_func(lambda rs: rs.outcome.result())\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 449, in result\n return self.__get_result()\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 401, in __get_result\n raise self._exception\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 130, in __call__\n result = await fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 94, in checked\n response = await raw(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 153, in raw\n return await original_query.__wrapped__(agent, *args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1128, in _query_llm\n llm_response = await chat.chat(\n ^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/llms/chat.py\", line 89, in chat\n llm_response: LLMResponse = await self._model.call(\n ^^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 145, in call\n return await original_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 232, in async_wrapped\n return await copy(fn, *args, **kwargs) # type: ignore[type-var]\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 126, in __call__\n do = await self.iter(retry_state=retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 171, in iter\n result = await action(retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/_utils.py\", line 130, in inner\n return call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/__init__.py\", line 462, in <lambda>\n self._add_action_func(lambda rs: rs.outcome.result())\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 449, in result\n return self.__get_result()\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 401, in __get_result\n raise self._exception\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 130, in __call__\n result = await fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/llms/lite_llm.py\", line 370, in call\n response = await litellm.acompletion(**completion_kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/utils.py\", line 1689, in wrapper_async\n result = await original_function(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/main.py\", line 643, in acompletion\n response = await init_response\n ^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/openai/openai.py\", line 886, in acompletion\n headers, response = await self.make_openai_chat_completion_request(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/litellm_core_utils/logging_utils.py\", line 289, in async_wrapper\n result = await func(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/openai/openai.py\", line 427, in make_openai_chat_completion_request\n raw_response = await openai_aclient.chat.completions.with_raw_response.create(**data, timeout=timeout)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_legacy_response.py\", line 386, in wrapped\n return cast(LegacyAPIResponse[R], await func(*args, **kwargs))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/resources/chat/completions/completions.py\", line 2907, in create\n return await self._post(\n ^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_base_client.py\", line 1992, in post\n return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_base_client.py\", line 1709, in request\n response = await self._send_request(\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_client.py\", line 1097, in _send_request\n response = await self._send_with_auth_retry(request, stream=stream, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_client.py\", line 1075, in _send_with_auth_retry\n response = await super()._send_request(request, stream=stream, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_base_client.py\", line 1628, in _send_request\n return await self._client.send(request, stream=stream, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1629, in send\n response = await self._send_handling_auth(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1657, in _send_handling_auth\n response = await self._send_handling_redirects(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1694, in _send_handling_redirects\n response = await self._send_single_request(request)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1730, in _send_single_request\n response = await transport.handle_async_request(request)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/custom_httpx/aiohttp_transport.py\", line 310, in handle_async_request\n response = await self._make_aiohttp_request(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/custom_httpx/aiohttp_transport.py\", line 288, in _make_aiohttp_request\n response = await client_session.request(**request_kwargs).__aenter__()\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client.py\", line 1756, in __aenter__\n self._resp: _RetType_co = await self._coro\n ^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client.py\", line 900, in _request\n resp = await handler(req)\n ^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client.py\", line 878, in _connect_and_send_request\n await resp.start(conn)\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client_reqrep.py\", line 745, in start\n message, payload = await protocol.read() # type: ignore[union-attr]\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/streams.py\", line 716, in read\n await self._waiter\nasyncio.exceptions.CancelledError\n\nThe above exception was the direct cause of the following exception:\n\nTraceback (most recent call last):\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/trial/trial.py\", line 554, in _run_agent_phase\n await asyncio.wait_for(\n File \"/usr/local/lib/python3.12/asyncio/tasks.py\", line 519, in wait_for\n async with timeouts.timeout(timeout):\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/asyncio/timeouts.py\", line 115, in __aexit__\n raise TimeoutError from exc_val\nTimeoutError\n\nThe above exception was the direct cause of the following exception:\n\nTraceback (most recent call last):\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 180, in run\n await original(*args, **kwargs)\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/trial/trial.py\", line 563, in _run_agent_phase\n raise AgentTimeoutError(\nharbor.trial.errors.AgentTimeoutError: Agent execution timed out after 900.0 seconds\n", "occurred_at": "2026-11-16T09:32:20.815752Z"}, "original_result_sha256": "ad9874b6f77307a1f8f98a80a71f49169328cbc897f3245d794b60cf7d4e733d", "generation_or_verifier_rerun": false, "reason": "User-adjudicated zero for missing verdict; not a native completed verifier test verdict."}, "git-multibranch--attempt-002": {"method": "explicit_user_authorized_zero", "authorization": "[REDACTED_CREDENTIAL]", "timestamp": "2026-11-16T16:48:15.931448+00:00", "reward": 0.0, "native_reward": null, "original_verifier_result": null, "original_exception_info": {"exception_type": "AgentTimeoutError", "exception_message": "Agent execution timed out after 900.0 seconds", "exception_traceback": "Traceback (most recent call last):\n File \"/usr/local/lib/python3.12/asyncio/tasks.py\", line 520, in wait_for\n return await fut\n ^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1743, in run\n await self._run_agent_loop(\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1410, in _run_agent_loop\n ) = await self._handle_llm_interaction(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1288, in _handle_llm_interaction\n llm_response = await self._query_llm(\n ^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 111, in query\n response = await retry(checked, *args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 126, in __call__\n do = await self.iter(retry_state=retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 171, in iter\n result = await action(retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/_utils.py\", line 130, in inner\n return call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/__init__.py\", line 462, in <lambda>\n self._add_action_func(lambda rs: rs.outcome.result())\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 449, in result\n return self.__get_result()\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 401, in __get_result\n raise self._exception\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 130, in __call__\n result = await fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 94, in checked\n response = await raw(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 153, in raw\n return await original_query.__wrapped__(agent, *args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/agents/terminus_2/terminus_2.py\", line 1128, in _query_llm\n llm_response = await chat.chat(\n ^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/llms/chat.py\", line 89, in chat\n llm_response: LLMResponse = await self._model.call(\n ^^^^^^^^^^^^^^^^^^^^^^^\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 145, in call\n return await original_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 232, in async_wrapped\n return await copy(fn, *args, **kwargs) # type: ignore[type-var]\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 126, in __call__\n do = await self.iter(retry_state=retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 171, in iter\n result = await action(retry_state)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/_utils.py\", line 130, in inner\n return call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/__init__.py\", line 462, in <lambda>\n self._add_action_func(lambda rs: rs.outcome.result())\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 449, in result\n return self.__get_result()\n ^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/concurrent/futures/_base.py\", line 401, in __get_result\n raise self._exception\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/tenacity/asyncio/__init__.py\", line 130, in __call__\n result = await fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/llms/lite_llm.py\", line 370, in call\n response = await litellm.acompletion(**completion_kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/utils.py\", line 1689, in wrapper_async\n result = await original_function(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/main.py\", line 643, in acompletion\n response = await init_response\n ^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/openai/openai.py\", line 886, in acompletion\n headers, response = await self.make_openai_chat_completion_request(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/litellm_core_utils/logging_utils.py\", line 289, in async_wrapper\n result = await func(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/openai/openai.py\", line 427, in make_openai_chat_completion_request\n raw_response = await openai_aclient.chat.completions.with_raw_response.create(**data, timeout=timeout)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_legacy_response.py\", line 386, in wrapped\n return cast(LegacyAPIResponse[R], await func(*args, **kwargs))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/resources/chat/completions/completions.py\", line 2907, in create\n return await self._post(\n ^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_base_client.py\", line 1992, in post\n return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_base_client.py\", line 1709, in request\n response = await self._send_request(\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_client.py\", line 1097, in _send_request\n response = await self._send_with_auth_retry(request, stream=stream, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_client.py\", line 1075, in _send_with_auth_retry\n response = await super()._send_request(request, stream=stream, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/openai/_base_client.py\", line 1628, in _send_request\n return await self._client.send(request, stream=stream, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1629, in send\n response = await self._send_handling_auth(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1657, in _send_handling_auth\n response = await self._send_handling_redirects(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1694, in _send_handling_redirects\n response = await self._send_single_request(request)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/httpx/_client.py\", line 1730, in _send_single_request\n response = await transport.handle_async_request(request)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/custom_httpx/aiohttp_transport.py\", line 310, in handle_async_request\n response = await self._make_aiohttp_request(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/litellm/llms/custom_httpx/aiohttp_transport.py\", line 288, in _make_aiohttp_request\n response = await client_session.request(**request_kwargs).__aenter__()\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client.py\", line 1756, in __aenter__\n self._resp: _RetType_co = await self._coro\n ^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client.py\", line 900, in _request\n resp = await handler(req)\n ^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client.py\", line 878, in _connect_and_send_request\n await resp.start(conn)\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/client_reqrep.py\", line 745, in start\n message, payload = await protocol.read() # type: ignore[union-attr]\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/aiohttp/streams.py\", line 716, in read\n await self._waiter\nasyncio.exceptions.CancelledError\n\nThe above exception was the direct cause of the following exception:\n\nTraceback (most recent call last):\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/trial/trial.py\", line 554, in _run_agent_phase\n await asyncio.wait_for(\n File \"/usr/local/lib/python3.12/asyncio/tasks.py\", line 519, in wait_for\n async with timeouts.timeout(timeout):\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/usr/local/lib/python3.12/asyncio/timeouts.py\", line 115, in __aexit__\n raise TimeoutError from exc_val\nTimeoutError\n\nThe above exception was the direct cause of the following exception:\n\nTraceback (most recent call last):\n File \"/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/code/eval-framework/scripts/agentic_recovery.py\", line 180, in run\n await original(*args, **kwargs)\n File \"/opt/scicoding/venvs/agentic-qualified-20261109/lib/python3.12/site-packages/harbor/trial/trial.py\", line 563, in _run_agent_phase\n raise AgentTimeoutError(\nharbor.trial.errors.AgentTimeoutError: Agent execution timed out after 900.0 seconds\n", "occurred_at": "2026-11-16T10:12:35.491721Z"}, "original_result_sha256": "7730cb4b57f22d82c0f3206561df48c480acf18e6eff65fd712ba26207d5b4ec", "generation_or_verifier_rerun": false, "reason": "User-adjudicated zero for missing verdict; not a native completed verifier test verdict."}}
|
granite-4.2-8b/terminal-bench-2.1/scoring_adjudications.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dd410307a0fb34f48cec7a0a7ce15bbcc34d390fb2207ee8c8873c29518f561
|
| 3 |
+
size 46746
|
granite-4.2-8b/terminal-bench-2.1/summary.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"tasks": 89, "attempts": 178, "attempts_per_task": 2, "metric": "avg@2", "passed": 32, "avg_at_2": 0.1797752808988764, "unscored": 0, "agent_timeouts":
|
|
|
|
| 1 |
+
{"tasks": 89, "attempts": 178, "attempts_per_task": 2, "metric": "avg@2", "passed": 32, "avg_at_2": 0.1797752808988764, "unscored": 0, "agent_timeouts": 103, "reward_positive_with_exception": 3, "scoring": "native_verifier_with_explicit_user_adjudication", "scoring_policy": "native-verifier-after-agent-stop-v1", "generation_finished": true, "grading_complete": true, "provisional": false, "source_run": "/workspace/data/sci-llm-training/runtime/agentic-granite-oss-followups-20261115/01-granite42-8b", "sampling": {"temperature": 0.7, "top_p": 0.95, "top_k": 20}, "hardware": "H100", "tasks_per_engine": 2, "adjudicated_zero": 2, "native_scored": 176}
|
granite-4.2-8b/terminal-bench-2.1/trajectories.jsonl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9804c92f45b541a09258f524573ef8e7e8c2f9d6550a6c5759030937db83d4a5
|
| 3 |
+
size 63305021
|