narrative string | question string | choices string | answer_index int64 | answer_choice string | task_source string | formatted_prompt list | response list | formatted_prompt__metadata list | response__metadata list | eval_result list | eval_result__metadata list | eval_set string | eval_set_label string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
In the tranquil solitude of a secluded chalet, Clyde meets an untimely death from snake venom; a case which sees the sharp mind of Detective Winston matching wits with two elusive suspects, Mack and Taylor.
Detective Winston was looking at the tailor-stitched files bundled together with papers on a desk in his office.... | Who is the most likely murderer? | ['Mack', 'Taylor'] | 0 | Mack | musr | [
{
"content": "You are a helpful assistant that will answer the questions given by the user.\n\nIMPORTANT DOMAIN KNOWLEDGE - Use these facts to guide your reasoning about murder mystery problems:\n\n 1. Always list and evaluate means, motive, and opportunity for every suspect—note which of the three is strong, ... | [
"To determine the most likely murderer, we need to evaluate the means, motive, and opportunity for both suspects.\n\n**Means:**\nTo establish means, we need to determine if the suspect had access to the snake venom. \n\n- **Taylor:** As a reptile handler and owner of a cleaning business, Taylor has frequent contact... | [
{
"endpoint_type": "chat",
"litellm_function": "acompletion",
"model": "together_ai/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo",
"raw_prompt": [
{
"content": "You are a helpful assistant that will answer the questions given by the user.\n\nIMPORTANT DOMAIN KNOWLEDGE - Use these facts ... | [
{
"cost": 0.00045845999999999994,
"finish_reason": "stop",
"input_tokens": 2057,
"model": "together_ai/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo",
"output_tokens": 490,
"raw_response": "{\"id\": \"oYcH4i9-4YNCb4-9d2c1a95d88b8c3f\", \"created\": 1771906128, \"model\": \"together_ai/meta-lla... | [
false,
false,
true,
false
] | [
{
"correct_answer_index": 0,
"correct_answer_number": 1,
"correct_answer_text": "Mack",
"extracted_number": 2,
"extracted_text": "2",
"extractor": "musr",
"matched_by": "number",
"num_choices": 2,
"raw_line": "ANSWER: 2"
},
{
"correct_answer_index": 0,
"correct_answer... | original_enhanced | Original (enhanced prompt) |
"In the dim light of an abandoned factory, Kaylee's lifeless body lay with the cold steel of a picka(...TRUNCATED) | Who is the most likely murderer? | ['Penelope', 'Francis'] | 0 | Penelope | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, we need to analyze the evidence for both Penelope and Franc(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.0004671,"finish_reason":"stop","input_tokens":2112,"model":"together_ai/meta-llama/Meta-L(...TRUNCATED) | [
true,
true,
true,
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Penelope","extracted_num(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In a shocking twist of events, Charlotte is found lifeless in her kitchen, felled by a stun gun, an(...TRUNCATED) | Who is the most likely murderer? | ['Letti', 'Deborah'] | 0 | Letti | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, I will evaluate each suspect based on the criteria of motiv(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00046926,"finish_reason":"stop","input_tokens":2147,"model":"together_ai/meta-llama/Meta-(...TRUNCATED) | [
true,
true,
true,
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Letti","extracted_number(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In the high-intensity world of a local fitness center, Tommy's exercise regime is abruptly terminat(...TRUNCATED) | Who is the most likely murderer? | ['Lucy', 'Jacqueline'] | 0 | Lucy | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, we need to analyze each suspect's means, motive, and opport(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.0004617,"finish_reason":"stop","input_tokens":2081,"model":"together_ai/meta-llama/Meta-L(...TRUNCATED) | [
true,
false,
false,
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Lucy","extracted_number"(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In a chilling turn of events, Rose is found lifeless in a car, killed by a vial of acid, leading De(...TRUNCATED) | Who is the most likely murderer? | ['Dexter', 'Daisy'] | 1 | Daisy | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, I will evaluate each suspect based on the criteria of motiv(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00046385999999999997,"finish_reason":"stop","input_tokens":2087,"model":"together_ai/meta(...TRUNCATED) | [
true,
true,
true,
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Daisy","extracted_number(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In the dense, dank heart of the rainforest, Skylar's life was brutally silenced with a lead pipe, l(...TRUNCATED) | Who is the most likely murderer? | ['Albert', 'Mya'] | 1 | Mya | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, we need to evaluate the means, motive, and opportunity for (...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00047646,"finish_reason":"stop","input_tokens":2183,"model":"together_ai/meta-llama/Meta-(...TRUNCATED) | [
true,
false,
false,
false
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Mya","extracted_number":(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In the hallowed silence of a synagogue, a shocking murder unfolds as Derek is impaled by a harpoon,(...TRUNCATED) | Who is the most likely murderer? | ['Hailey', 'Angela'] | 1 | Angela | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, we need to evaluate the means, motive, and opportunity for (...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00051066,"finish_reason":"stop","input_tokens":2443,"model":"together_ai/meta-llama/Meta-(...TRUNCATED) | [
true,
true,
true,
false
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Angela","extracted_numbe(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"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 | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, let's analyze the evidence and evaluate the means, motive, (...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00048726,"finish_reason":"stop","input_tokens":2244,"model":"together_ai/meta-llama/Meta-(...TRUNCATED) | [
false,
false,
false,
false
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Rosemary","extracted_num(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In the aftermath of a devastating home explosion that claims the life of Angela, seasoned Detective(...TRUNCATED) | Who is the most likely murderer? | ['Patrick', 'Alvin'] | 0 | Patrick | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, we need to analyze the means, motive, and opportunity for b(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00047646,"finish_reason":"stop","input_tokens":2192,"model":"together_ai/meta-llama/Meta-(...TRUNCATED) | [
true,
false,
true,
true
] | [{"correct_answer_index":0,"correct_answer_number":1,"correct_answer_text":"Patrick","extracted_numb(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
"In the frost-kissed tranquility of a chalet, Malaysia meets a fatal end by a rifle's sinister call,(...TRUNCATED) | Who is the most likely murderer? | ['Letti', 'Evelyn'] | 1 | Evelyn | musr | [{"content":"You are a helpful assistant that will answer the questions given by the user.\n\nIMPORT(...TRUNCATED) | ["To determine the most likely murderer, I will evaluate each suspect based on the means, motive, an(...TRUNCATED) | [{"endpoint_type":"chat","litellm_function":"acompletion","model":"together_ai/meta-llama/Meta-Llama(...TRUNCATED) | [{"cost":0.00044406,"finish_reason":"stop","input_tokens":2011,"model":"together_ai/meta-llama/Meta-(...TRUNCATED) | [
false,
false,
false,
true
] | [{"correct_answer_index":1,"correct_answer_number":2,"correct_answer_text":"Evelyn","extracted_numbe(...TRUNCATED) | original_enhanced | Original (enhanced prompt) |
End of preview. Expand in Data Studio
t1-musr-prompt-enhancement-together_ai-meta-llama-meta-llama-3-1-8b-s5-enhanced
Phase 5: enhanced eval (3 eval sets)
Dataset Info
- Rows: 150
- Columns: 14
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 |
| task_source | Value('string') | No description provided |
| formatted_prompt | List({'content': Value('string'), 'role': 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': Value('string')}) | 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': Value('string')}) | 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 |
| eval_set | Value('string') | No description provided |
| eval_set_label | Value('string') | No description provided |
Generation Parameters
{
"script_name": "musr_prompt_enhancement/run_experiment.py",
"model": "together_ai/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo",
"description": "Phase 5: enhanced eval (3 eval sets)",
"custom_metadata": {
"stage": "s5_enhanced",
"eval_sets": {
"original_enhanced": {
"total_problems": 50,
"samples_per_problem": 4,
"pass@1": 0.73,
"pass@2": 0.8400000000000002,
"pass@3": 0.875,
"pass@4": 0.9
},
"heldout_base": {
"total_problems": 50,
"samples_per_problem": 4,
"pass@1": 0.705,
"pass@2": 0.8433333333333333,
"pass@3": 0.9,
"pass@4": 0.94
},
"heldout_enhanced": {
"total_problems": 50,
"samples_per_problem": 4,
"pass@1": 0.705,
"pass@2": 0.8399999999999999,
"pass@3": 0.9,
"pass@4": 0.94
}
}
},
"hyperparameters": {},
"input_datasets": []
}
Usage
from datasets import load_dataset
dataset = load_dataset("reasoning-degeneration-dev/t1-musr-prompt-enhancement-together_ai-meta-llama-meta-llama-3-1-8b-s5-enhanced", split="train")
print(f"Loaded {len(dataset)} rows")
This dataset is tracked in reasoning-degeneration-dev/PROJECT-MANIFEST
- Downloads last month
- 12