narrative string | question string | choices string | answer_index int64 | answer_choice string | formatted_prompt list | task_source string | domain string | response list | formatted_prompt__metadata list | response__metadata 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": "Forget everything you know about detective frameworks. Do NOT think in terms of \"means, motive, and opportunity\" — those categories will not help you here and you should ... | musr | murder_mysteries | [
"<think>\nOkay, let's try to figure this out. So, the story is about a murder at a bungee jumping site where Mack was killed with a nunchaku. Detective Winston is investigating, and we need to determine who did it between Mackenzie and Ana.\n\nFirst, I need to reconstruct the timeline step by step. Let's go through... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Thinking",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.",
"role": "system"
},
{
"conten... | [
{
"cost": 0.00801435,
"finish_reason": "stop",
"input_tokens": 1549,
"model": "together_ai/Qwen/Qwen3-Next-80B-A3B-Thinking",
"output_tokens": 5188,
"raw_response": {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"con... | [
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 e(...TRUNCATED) | 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":"(...TRUNCATED) | musr | murder_mysteries | ["<think>\nOkay, let's tackle this step by step. I need to reconstruct the timeline of events from t(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.005539049999999999,"finish_reason":"stop","input_tokens":1497,"model":"together_ai/Qwen/Q(...TRUNCATED) | [
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Ana","extracted_number":(...TRUNCATED) |
"In the haze of neon lights and the serving of a silent hand of fate, Timothy lies dead in a casino,(...TRUNCATED) | 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":"(...TRUNCATED) | musr | murder_mysteries | ["<think>\nOkay, let's tackle this step by step. The user wants me to forget detective frameworks li(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.00846675,"finish_reason":"stop","input_tokens":1505,"model":"together_ai/Qwen/Qwen3-Next-(...TRUNCATED) | [
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Harry","extracted_number(...TRUNCATED) |
"In the haze of neon lights and the serving of a silent hand of fate, Timothy lies dead in a casino,(...TRUNCATED) | 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":"(...TRUNCATED) | musr | murder_mysteries | ["<think>\nOkay, let's try to figure out who the murderer is here. The story says Timothy was killed(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.009294,"finish_reason":"stop","input_tokens":1610,"model":"together_ai/Qwen/Qwen3-Next-80(...TRUNCATED) | [
false
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Rosemary","extracted_num(...TRUNCATED) |
"In the quiet chaos of the city zoo, Detective Winston finds himself investigating an unusual murder(...TRUNCATED) | 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":"(...TRUNCATED) | musr | murder_mysteries | ["<think>\nFirst, I need to forget about \"means, motive, and opportunity\" as categories. I shouldn(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.01720755,"finish_reason":"stop","input_tokens":1437,"model":"together_ai/Qwen/Qwen3-Next-(...TRUNCATED) | [
false
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Amelia","extracted_numbe(...TRUNCATED) |
"In the quiet chaos of the city zoo, Detective Winston finds himself investigating an unusual murder(...TRUNCATED) | 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":"(...TRUNCATED) | musr | murder_mysteries | ["<think>\nOkay, let's try to figure out who the murderer is here. The story is about Daniel, a zook(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.01343085,"finish_reason":"stop","input_tokens":1529,"model":"together_ai/Qwen/Qwen3-Next-(...TRUNCATED) | [
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Emma","extracted_number"(...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'] | 0 | Terrance | [{"content":"You are a helpful assistant that will answer the questions given by the user.","role":"(...TRUNCATED) | musr | murder_mysteries | ["<think>\nOkay, let's tackle this step by step. The user wants me to forget detective frameworks li(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.01006605,"finish_reason":"stop","input_tokens":1477,"model":"together_ai/Qwen/Qwen3-Next-(...TRUNCATED) | [
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 | ["<think>\nOkay, let's see. I need to figure out who the murderer is based on the story. The user sa(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.0195975,"finish_reason":"stop","input_tokens":1230,"model":"together_ai/Qwen/Qwen3-Next-8(...TRUNCATED) | [
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 | ["<think>\nOkay, let me try to figure this out. So the story is about Kent being killed by snake ven(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.011198099999999999,"finish_reason":"stop","input_tokens":1364,"model":"together_ai/Qwen/Q(...TRUNCATED) | [
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 | ["<think>\nOkay, let's tackle this step by step. So, the task is to reconstruct the timeline of even(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/Qwen/Qwen3-Next-80B-A(...TRUNCATED) | [{"cost":0.01390275,"finish_reason":"stop","input_tokens":1345,"model":"together_ai/Qwen/Qwen3-Next-(...TRUNCATED) | [
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Julia","extracted_number(...TRUNCATED) |
t1-musr-timeline-murder-together_ai-qwen-qwen3-next-80b-a3b-thin-a282ebac
Timeline reconstruction evaluation for MuSR murder mysteries. Completely abandons the means/motive/opportunity framework. Instead, the model reconstructs the chronological sequence of events, traces character movements, and determines the murderer by following the story to its logical conclusion.
Stylistically, the model uses narrative storytelling language: "So first we see...", "Then what's interesting is...", "Now here's the key part...".
Comparison to Baseline
| Metric | Baseline (cot+) | Timeline | Delta |
|---|---|---|---|
| pass@1 | 0.8000 | 0.8000 | +0.0000 |
Results
| Metric | Value |
|---|---|
| pass@1 | 0.8000 |
| Total problems | 10 |
Prompt (excerpt)
Forget everything you know about detective frameworks. Do NOT think in terms of "means, motive, and opportunity" — those categories will not help you here and you should not use them.
Instead, reconstruct what happened. Read the story and build a chronological timeline of events.
Narrate the timeline as if you're retelling the events to someone who hasn't read the story. Use natural storytelling language:
- "So first we learn that..."
- "Then what happens is..."
- "And this is where things get interesting, because..."
- "Now here's the key part..."
- "What's really telling is that..."
- "So if we put all of this together..."
Go through the story from beginning to end. For each event, note:
- What happened
- Who was involved
- Where it took place
- What it reveals about people's situations and relationships
After reconstructing the full timeline, step back and tell me: given everything that happened in sequence, who is the only person who could have done this? Don't categorize or organize evidence — just follow the story to its natural conclusion. The narrative itself will tell you who did it.
You must pick one suspect....
Details
| Parameter | Value |
|---|---|
| Model | together_ai/Qwen/Qwen3-Next-80B-A3B-Thinking |
| Temperature | 0.7 |
| Top-p | 0.95 |
| Max tokens | 32768 |
| Samples | 1 |
| Thinking | True |
Usage
from datasets import load_dataset
dataset = load_dataset("reasoning-degeneration-dev/t1-musr-timeline-murder-together_ai-qwen-qwen3-next-80b-a3b-thin-a282ebac", split="train")
Tracked in reasoning-degeneration-dev/PROJECT-MANIFEST
- Downloads last month
- 34