narrative string | question string | choices string | answer_index int64 | answer_choice string | formatted_prompt list | task_source string | domain string | prompt_variant string | response list | formatted_prompt__metadata list | response__metadata list | wingdings_compliance list | eval_result list | eval_result__metadata list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
In an adrenaline inducing bungee jumping site, Mack's thrill-seeking adventure came to a gruesome end by a nunchaku; now, it's up to Detective Winston to unravel the deadly secrets between Mackenzie and Ana.
Winston took a gulp of his black coffee, staring at the notes sprawled across his desk. A murder case at a bung... | Who is the most likely murderer? | ['Mackenzie', 'Ana'] | 0 | Mackenzie | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"content": "You are a detective solving a murder mystery. You must determine which suspect is the murderer.\n\nThe murderer needs MEANS (access to weapon), MOTIVE (reason to kill), and... | musr | murder_mysteries | wingdings | [
"β Ana: βββ | β₯ββ | βββ \nβ Mackenzie: βββ | β₯ββ | βββ \n\nβββ (nunchaku purchase + martial arts training) \nβ₯ββ (jealousy over promotion + public arguments + emotional rivalry) \nβββ (saw at scene + no alibi + weapon matched) \n\nAna: \nβββ (no access to weapon | no record of nunchaku | cleaning cloth β weap... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.0005124,
"finish_reason": "stop",
"input_tokens": 1606,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"output_tokens": 181,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"conte... | [
0.22025316455696198
] | [
true
] | [
{
"correct_answer_index": 0,
"correct_answer_number": 1,
"correct_answer_text": "Mackenzie",
"extracted_number": 1,
"extracted_text": "1",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 1"
}
] |
In an adrenaline inducing bungee jumping site, Mack's thrill-seeking adventure came to a gruesome end by a nunchaku; now, it's up to Detective Winston to unravel the deadly secrets between Mackenzie and Ana.
Winston sat quietly in an unmarked Ford, parked a block away from Mackenzie's downtown apartment building. He o... | Who is the most likely murderer? | ['Mackenzie', 'Ana'] | 1 | Ana | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"content": "You are a detective solving a murder mystery. You must determine which suspect is the murderer.\n\nThe murderer needs MEANS (access to weapon), MOTIVE (reason to kill), and... | musr | murder_mysteries | wingdings | [
"β Mackenzie: βββ (nunchaku purchase + training) | β₯ββ (promotion + cufflinks envy) | βββ (at apartment all day) \nβ Ana: βββ (nunchaku at home + training at site) | β₯ββ (ridicule + belief disrespected) | βββ (at scene with victim) \n\nβ‘ Mackenzie: βββ | β₯ββ | βββ β΄ β \nβ‘ Ana: βββ | β₯ββ | βββ β΄ β
ALL THREE β
\n... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.00047159999999999997,
"finish_reason": "stop",
"input_tokens": 1554,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"output_tokens": 159,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
... | [
0.2666666666666667
] | [
true
] | [
{
"correct_answer_index": 1,
"correct_answer_number": 2,
"correct_answer_text": "Ana",
"extracted_number": 2,
"extracted_text": "2",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 2"
}
] |
In the haze of neon lights and the serving of a silent hand of fate, Timothy lies dead in a casino, a sai his cruel end, leaving the unruffled Detective Winston to interrogate suspects, Harry and Rosemary.
It had been a long day for Winston. The air was heavy with the scent of fresh coffee and the clamour of a bustlin... | Who is the most likely murderer? | ['Harry', 'Rosemary'] | 0 | Harry | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"content": "You are a detective solving a murder mystery. You must determine which suspect is the murderer.\n\nThe murderer needs MEANS (access to weapon), MOTIVE (reason to kill), and... | musr | murder_mysteries | wingdings | [
"β Harry: βββ | β₯ββ | βββ \nβ Rosemary: βββ | β₯ββ | βββ \n\nβ‘ Harry: βββ (sai in home, practiced daily) \nβ‘ Harry: β₯ββ (debt β threat of exposure β motive) \nβ‘ Harry: βββ (casino footage, VIP lounge access) \n\nβ‘ Rosemary: βββ (collection includes sais) \nβ‘ Rosemary: β₯ββ (racial hostility, public threats) \n... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.0006003,
"finish_reason": "stop",
"input_tokens": 1562,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"output_tokens": 244,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"conte... | [
0.23274161735700194
] | [
true
] | [
{
"correct_answer_index": 0,
"correct_answer_number": 1,
"correct_answer_text": "Harry",
"extracted_number": 1,
"extracted_text": "1",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 1"
}
] |
In the haze of neon lights and the serving of a silent hand of fate, Timothy lies dead in a casino, a sai his cruel end, leaving the unruffled Detective Winston to interrogate suspects, Harry and Rosemary.
Detective Winston wearily walked into the noisy, packed casino, the glow of the neon lights casting an unnatural ... | Who is the most likely murderer? | ['Harry', 'Rosemary'] | 1 | Rosemary | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"content": "You are a detective solving a murder mystery. You must determine which suspect is the murderer.\n\nThe murderer needs MEANS (access to weapon), MOTIVE (reason to kill), and... | musr | murder_mysteries | wingdings | [
"β Harry: βββ | β₯ββ | βββ \nβ Rosemary: βββ | β₯ββ | βββ \n\nβ‘ Harry: βββ (sai mastery + daily practice) \nβ‘ Harry: β₯ββ (debt β motive) \nβ‘ Harry: βββ (cam footage β entered before murder) \nβ‘ Harry: β alias use β β concealment β β alibi? β β no alibi \n\nβ‘ Rosemary: βββ (weapon collection β access) \nβ‘ Rosem... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.0007255499999999999,
"finish_reason": "stop",
"input_tokens": 1667,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"output_tokens": 317,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
... | [
0.209563994374121
] | [
false
] | [
{
"correct_answer_index": 1,
"correct_answer_number": 2,
"correct_answer_text": "Rosemary",
"extracted_number": 1,
"extracted_text": "1",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 1"
}
] |
In the quiet chaos of the city zoo, Detective Winston finds himself investigating an unusual murder, where Daniel, a zookeeper, was fatally injected by a syringe, with only two suspects in sight: Amelia, the grieving widow, and Emma, the secretive co-worker.
Detective Winston stood in the waiting room of the hospital ... | Who is the most likely murderer? | ['Amelia', 'Emma'] | 0 | Amelia | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"content": "You are a detective solving a murder mystery. You must determine which suspect is the murderer.\n\nThe murderer needs MEANS (access to weapon), MOTIVE (reason to kill), and... | musr | murder_mysteries | wingdings | [
"β Amelia: βββ (bar syringes) | β₯ββ (humiliating photos) | βββ (zoo setup, but no alibi for time of death) \nβ Emma: βββ (hospital access, syringe skills) | β₯ββ (overacted grief, unknown relation) | βββ (organized zoo trip, calendar in apartment)\n\nβ‘ Amelia: βββ | β₯ββ | βββ β β ONE MISSING \nβ‘ Emma: βββ | β₯ββ | ... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.0007041,
"finish_reason": "stop",
"input_tokens": 1494,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"output_tokens": 320,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"conte... | [
0.161849710982659
] | [
false
] | [
{
"correct_answer_index": 0,
"correct_answer_number": 1,
"correct_answer_text": "Amelia",
"extracted_number": 2,
"extracted_text": "2",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 2"
}
] |
In the quiet chaos of the city zoo, Detective Winston finds himself investigating an unusual murder, where Daniel, a zookeeper, was fatally injected by a syringe, with only two suspects in sight: Amelia, the grieving widow, and Emma, the secretive co-worker.
Winston sat in his chair, sifting through files, his fingert... | Who is the most likely murderer? | ['Amelia', 'Emma'] | 1 | Emma | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"content": "You are a detective solving a murder mystery. You must determine which suspect is the murderer.\n\nThe murderer needs MEANS (access to weapon), MOTIVE (reason to kill), and... | musr | murder_mysteries | wingdings | [
"β Amelia: βββ (syringeβzoo/monkey) | β₯ββ (barβDanielβdebt?) | βββ (zooβwitnessed) \nβ Emma: βββ (nurseβaccess) | β₯ββ (lawsuitβtestimonyβcourt) | βββ (zooβplanned tripβseen) \n\nβ‘ Amelia: βββ
(syringe used on monkeyβsame tool?) \nβ‘ Emma: βββ
(syringe standard in hospitalβno unique trace) \n\nβ£ Amelia: β₯ββ¦ (pain... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.0008889,
"finish_reason": "stop",
"input_tokens": 1586,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"output_tokens": 434,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"conte... | [
0.19484240687679089
] | [
true
] | [
{
"correct_answer_index": 1,
"correct_answer_number": 2,
"correct_answer_text": "Emma",
"extracted_number": 2,
"extracted_text": "2",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 2"
}
] |
"In the grim spectacle of the wrestling ring stained with blood, Detective Winston embarks on a chil(...TRUNCATED) | Who is the most likely murderer? | ['Terrance', 'Nelson'] | 0 | Terrance | [{"content":"You are a helpful assistant that will answer the questions given by the user.","role":"(...TRUNCATED) | musr | murder_mysteries | wingdings | ["β Terrance: βββ | β₯ββ | βββ \nβ Nelson: βββ | β₯ββ | βββ (...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.0006711,"finish_reason":"stop","input_tokens":1534,"model":"together_ai/Qwen/Qwen3-Next-8(...TRUNCATED) | [
0.17871759890859484
] | [
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Terrance","extracted_num(...TRUNCATED) |
"In the grim spectacle of the wrestling ring stained with blood, Detective Winston embarks on a chil(...TRUNCATED) | Who is the most likely murderer? | ['Terrance', 'Nelson'] | 1 | Nelson | [{"content":"You are a helpful assistant that will answer the questions given by the user.","role":"(...TRUNCATED) | musr | murder_mysteries | wingdings | ["β Terrance: βββ | β₯ββ | βββ \nβ Nelson: βββ | β₯ββ | βββ (...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.0005965499999999999,"finish_reason":"stop","input_tokens":1287,"model":"together_ai/Qwen/(...TRUNCATED) | [
0.25794392523364484
] | [
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Nelson","extracted_numbe(...TRUNCATED) |
"In the hushed sanctity of a mosque, Kent falls victim to a deadly snake venom, leaving Detective Wi(...TRUNCATED) | Who is the most likely murderer? | ['Ronald', 'Julia'] | 0 | Ronald | [{"content":"You are a helpful assistant that will answer the questions given by the user.","role":"(...TRUNCATED) | musr | murder_mysteries | wingdings | ["β Ronald: βββ (keys to enclosures) | β₯ββ (indictments β motive) | βββ (guard(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.00046515,"finish_reason":"stop","input_tokens":1421,"model":"together_ai/Qwen/Qwen3-Next-(...TRUNCATED) | [
0.23268698060941828
] | [
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Ronald","extracted_numbe(...TRUNCATED) |
"In the hushed sanctity of a mosque, Kent falls victim to a deadly snake venom, leaving Detective Wi(...TRUNCATED) | Who is the most likely murderer? | ['Ronald', 'Julia'] | 1 | Julia | [{"content":"You are a helpful assistant that will answer the questions given by the user.","role":"(...TRUNCATED) | musr | murder_mysteries | wingdings | ["β Julia: βββ | β₯ββ | βββ \nβ Ronald: βββ | β₯ββ | βββ \n\(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.0005868,"finish_reason":"stop","input_tokens":1402,"model":"together_ai/Qwen/Qwen3-Next-8(...TRUNCATED) | [
0.23207547169811316
] | [
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Julia","extracted_number(...TRUNCATED) |
t1-wingdings-musr-murder-together_ai-qwen-qwen3-next-80b-a3b-inst-079fe525
Wingdings compliance eval on MuSR murder mysteries (instruct model)
Dataset Info
- Rows: 10
- Columns: 15
Columns
| Column | Type | Description |
|---|---|---|
| narrative | Value('string') | No description provided |
| question | Value('string') | No description provided |
| choices | Value('string') | No description provided |
| answer_index | Value('int64') | No description provided |
| answer_choice | Value('string') | No description provided |
| formatted_prompt | List({'content': Value('string'), 'role': Value('string')}) | No description provided |
| task_source | Value('string') | No description provided |
| domain | Value('string') | No description provided |
| prompt_variant | Value('string') | No description provided |
| response | List(Value('string')) | No description provided |
| formatted_prompt__metadata | List({'endpoint_type': Value('string'), 'litellm_function': Value('string'), 'model': Value('string'), 'raw_prompt': List({'content': Value('string'), 'role': Value('string')}), 'raw_system_prompt': Value('null'), 'request_payload': {'api_base': Value('string'), 'max_tokens': Value('int64'), 'messages': List({'content': Value('string'), 'role': Value('string')}), 'model': Value('string'), 'stream': Value('bool'), 'temperature': Value('float64'), 'timeout': Value('float64'), 'top_p': Value('float64')}}) | No description provided |
| response__metadata | List({'cost': Value('float64'), 'finish_reason': Value('string'), 'input_tokens': Value('int64'), 'model': Value('string'), 'output_tokens': Value('int64'), 'raw_response': {'choices': List({'finish_reason': Value('string'), 'index': Value('int64'), 'message': {'content': Value('string'), 'function_call': Value('null'), 'provider_specific_fields': {'reasoning': Value('null'), 'refusal': Value('null')}, 'role': Value('string'), 'tool_calls': Value('null')}, 'provider_specific_fields': {'seed': Value('null')}}), 'created': Value('int64'), 'id': Value('string'), 'metadata': {'weight_version': Value('string')}, 'model': Value('string'), 'object': Value('string'), 'prompt': List(Value('null')), 'service_tier': Value('null'), 'system_fingerprint': Value('null'), 'usage': {'completion_tokens': Value('int64'), 'completion_tokens_details': Value('null'), 'prompt_tokens': Value('int64'), 'prompt_tokens_details': Value('null'), 'total_tokens': Value('int64')}}}) | No description provided |
| wingdings_compliance | List(Value('float64')) | No description provided |
| eval_result | List(Value('bool')) | No description provided |
| eval_result__metadata | List({'correct_answer_index': Value('int64'), 'correct_answer_number': Value('int64'), 'correct_answer_text': Value('string'), 'extracted_number': Value('int64'), 'extracted_text': Value('string'), 'extractor': Value('string'), 'matched_by': Value('string'), 'num_choices': Value('int64'), 'raw_line': Value('string')}) | No description provided |
Generation Parameters
{
"script_name": "experiments/wingdings_compliance/musr_eval.py",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Instruct",
"description": "Wingdings compliance eval on MuSR murder mysteries (instruct model)",
"experiment": "wingdings_compliance",
"task": "musr_murder_mysteries",
"hyperparameters": {},
"input_datasets": []
}
Usage
from datasets import load_dataset
dataset = load_dataset("reasoning-degeneration-dev/t1-wingdings-musr-murder-together_ai-qwen-qwen3-next-80b-a3b-inst-079fe525", split="train")
print(f"Loaded {len(dataset)} rows")
This dataset is tracked in reasoning-degeneration-dev/PROJECT-MANIFEST
- Downloads last month
- 7