Nemotron-Post-Training-v3 Prompt-Only
Collection
51 items • Updated
Datasets:
dataset stringclasses 1
value | config stringclasses 1
value | split stringclasses 1
value | row_index int64 0 97k | prompt stringlengths 28 1.16k | prompt_source stringclasses 1
value | prompt_source_detail stringclasses 1
value | system_prompt stringclasses 406
values | system_source stringclasses 1
value | tools stringclasses 406
values | tools_source stringclasses 1
value | schema_str float64 | extraction_error float64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 0 |
Hi, I’m calling about the pre-authorization requirements for an elective procedure I’m planning that requires extended care coverage. Can you explain the insurance pre-authorization process for my situation and confirm if there are any coverage limitations for the services I’m planning? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Sen... | responses_create_params.input | [{"description": "Verify client identity using ID and birthdate", "name": "authenticate_client", "parameters": {"properties": {"birthdate": {"format": "date", "type": "string"}, "client_id": {"type": "string"}}, "required": ["client_id", "birthdate"]}, "strict": true, "type": "function"}, {"description": "Authenticate ... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 1 |
Hi, I’m reaching out because I noticed my recent order (DISC11) didn’t apply my SeasonPass loyalty discount. My Fan ID is LYL001, and I should qualify for the 15% tier benefit. Could you check if the discount can be applied retroactively to this purchase? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
### *... | responses_create_params.input | [{"description": "Retrieve fan profile details including loyalty tier, payment methods, and purchase history.", "name": "get_fan_profile", "parameters": {"properties": {"fan_id": {"description": "Unique fan identifier", "type": "string"}}, "required": ["fan_id"], "type": "object"}, "strict": true, "type": "function"}, ... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 2 |
Hi, I’m calling about the pre-authorization requirements for an elective procedure I’m planning that requires extended care coverage. Can you explain the insurance pre-authorization process for my situation and confirm if there are any coverage limitations for the services I’m planning? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Sen... | responses_create_params.input | [{"description": "Verify client identity using ID and birthdate", "name": "authenticate_client", "parameters": {"properties": {"birthdate": {"format": "date", "type": "string"}, "client_id": {"type": "string"}}, "required": ["client_id", "birthdate"]}, "strict": true, "type": "function"}, {"description": "Authenticate ... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 3 |
Hello, I'm a representative from the National Standardized Testing Board. I need to confirm whether your platform meets the continuing education requirements for professional licensing. Could you please provide official accreditation documentation or a letter verifying compliance with our standards? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Tes... | responses_create_params.input | [{"description": "Verify user identity using registered email address.", "name": "verify_email", "parameters": {"properties": {"email": {"description": "Email address to verify", "type": "string"}, "user_id": {"description": "Unique user identifier", "type": "string"}}, "required": ["user_id", "email"], "type": "object... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 4 |
Hello, I'm reaching out about our Herb Chopper that was previously used in a conventional kitchen. We're transitioning to organic production and need clarification on sanitation protocols. Could you explain how sanitation requirements differ between conventional and organic processing equipment? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Org... | responses_create_params.input | [{"description": "Verify user identity using:\n- Account ID + registered email\nOR\n- Order number + last 4 digits of payment method", "name": "authenticate_user", "parameters": {"properties": {"account_id": {"description": "User account ID (if using account_id_email method)", "title": "Account Id", "type": "string"}, ... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 5 |
Hello, I’m calling regarding my solar panel system (Project ID: P-778800). It’s been 7 years since installation, and I’ve noticed yellowing on 12 of the panels. I’d like to file a warranty claim for this issue. Could you let me know what steps I need to take to provide proof of proper maintenance (e.g., no chemical e... | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Ren... | responses_create_params.input | [{"description": "Verify customer ID and project ID association before accessing records. Ensures proper ownership and prevents unauthorized access.", "name": "verify_project_customer", "parameters": {"properties": {"customer_id": {"description": "Customer ID to verify", "type": "string"}, "project_id": {"description":... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 6 |
Hi, I'm a Pioneer member with ID P88990, last name Williams. I need to confirm if the "guaranteed guides" benefit is included for my tier. Can you check that for me? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Sum... | responses_create_params.input | [{"description": "Authenticate customer identity using approved methods: Booking ID + Email, Membership ID + Last Name, or Phone + Security Code", "name": "verify_identity", "parameters": {"properties": {"booking_id": {"type": "string"}, "email": {"type": "string"}, "last_name": {"type": "string"}, "membership_id": {"t... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 7 |
Hello, I'm having trouble receiving the password reset email for my account. My User ID is SF890123. I clicked the "Forgot Password" link but haven't seen the verification email in my inbox. Could you help me troubleshoot this? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
### *... | responses_create_params.input | [{"description": "Retrieve user profile details including subscription tier, payment methods, preferences, and linked devices.", "name": "get_user_profile", "parameters": {"properties": {"user_id": {"description": "Unique user identifier", "title": "User Id", "type": "string"}}, "required": ["user_id"], "title": "param... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 8 |
Hello, I'm a Silver member currently billed on June 10, 2025, and I'd like to downgrade to the Bronze membership immediately. However, the system is requiring a 30-day notice period for this change. I'm on a tight student budget and need the downgrade to take effect right away to save costs. Could you explain why the... | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Fle... | responses_create_params.input | [{"description": "Authenticates user using registered email address. Required before accessing account-sensitive operations.", "name": "authenticate_user", "parameters": {"properties": {"email": {"description": "User's registered email address", "type": "string"}}, "required": ["email"], "title": "parameters", "type": ... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 9 |
Hi, I'd like to schedule a maintenance appointment for my home service. My preferred time is Saturday at 10 AM at 606 Hilltop Dr. Can you check if a technician is available for a 90-minute window that day? Also, could you confirm if there are any emergency fees for weekend service? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Pro... | responses_create_params.input | [{"description": "Schedule a new service appointment with 90-minute time slots. Requires >24hr notice for modification-friendly bookings.", "name": "schedule_appointment", "parameters": {"properties": {"customer_id": {"description": "Unique customer identifier", "type": "string"}, "equipment_details": {"description": "... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 10 |
Hi, I'd like to schedule a maintenance appointment for my home service. My preferred time is Saturday at 10 AM at 606 Hilltop Dr. Can you check if a technician is available for a 90-minute window that day? Also, could you confirm if there are any emergency fees for weekend service? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Pro... | responses_create_params.input | [{"description": "Schedule a new service appointment with 90-minute time slots. Requires >24hr notice for modification-friendly bookings.", "name": "schedule_appointment", "parameters": {"properties": {"customer_id": {"description": "Unique customer identifier", "type": "string"}, "equipment_details": {"description": "... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 11 |
Hello, I’m Taylor Reed from Review-LEAK-12. I’m calling regarding a suspected concept leak in our unpublished proposal for the “Shadow Labyrinth” escape room. We recently noticed striking similarities between our design and a newly launched room by another firm. Can you confirm if your team has protocols in place to ... | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Esc... | responses_create_params.input | [{"description": "Authenticate client using:\n- Client ID + registered business email OR\n- Project ID + client's full name\nRequired before any account or project actions.", "name": "authenticate_client", "parameters": {"anyOf": [{"required": ["client_id", "business_email"]}, {"required": ["project_id", "client_full_n... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 12 |
Hello, I’m Taylor Reed from Review-LEAK-12. I’m calling regarding a suspected concept leak in our unpublished proposal for the “Shadow Labyrinth” escape room. We recently noticed striking similarities between our design and a newly launched room by another firm. Can you confirm if your team has protocols in place to ... | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Esc... | responses_create_params.input | [{"description": "Authenticate client using:\n- Client ID + registered business email OR\n- Project ID + client's full name\nRequired before any account or project actions.", "name": "authenticate_client", "parameters": {"anyOf": [{"required": ["client_id", "business_email"]}, {"required": ["project_id", "client_full_n... | responses_create_params.tools | null | null |
nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | default | train | 13 |
Hello, I'm calling to check if the Tesla Powerwall is compatible with my existing solar system. I currently have an 8kW setup with a SolarEdge inverter and monocrystalline panels. Can you help me with this? | responses_input | first_user_message | You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.
In each turn you can either:
- Send a message to the user.
- Make a tool call.
You cannot do both at the same time.
<policy>
# Sol... | responses_create_params.input | [{"description": "Verify user identity using account number and email or phone and last 4 SSN digits.", "name": "authenticate_user", "parameters": {"properties": {"account_number": {"description": "Required if method is 'account_email'", "title": "Account Number", "type": "string"}, "email": {"description": "Required i... | responses_create_params.tools | null | null |
End of preview. Expand in Data Studio
Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1-prompt-only
Prompt-only extraction from nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1.
Files:
prompts.csv: one prompt extraction record per source row. Records includeprompt, separatedsystem_prompt, and structuredtoolswhen the source row defines available tools. Nested values are JSON-encoded inside CSV cells.summary.md: source row counts, extracted row counts, count deltas, and failed prompt counts.null_or_empty_rows.md: row indexes where prompt extraction produced a null or empty prompt.
Summary:
- extracted rows: 96968
- failed prompt rows: 0
- row count delta: 0
Uploaded under jamesdborin from the Nemotron Post-Training v3 prompt extractor workflow.
- Downloads last month
- 102