Modelpublic

jastorj/couchmind-rlt-v5.8.2_rl_5K_ex-cw-11K-16bit

sourceHugging Faceapache-2.0updated 1mo agoView on Hugging Face

0likes12downloads

Model Card

jastorj/couchmind-v5.8.2coldstart-cw-26K-16bit Fine-tuned for NL2SQL++ v5.8.2rl5K_ex (RLT)

This model is a reinforcement-learning fine-tuned version of jastorj/couchmind-v5.8.2_cold_start-cw-26K-16bit on the NL2SQL++ v5.8.2rl5K_ex dataset.

Model Details

—Base Model: jastorj/couchmind-v5.8.2coldstart-cw-26K-16bit
—Task: Text-to-SQL generation
—Dataset: NL2SQL++ v5.8.2rl5K_ex
—Fine-tuning Method: GRPO (Group Relative Policy Optimisation) with Unsloth
—Quantization: 16-bit merged weights
—Training Dataset Size: 5021 examples
—Validation Dataset Size: 0 examples

Full training configuration: see training_args.json in this repo.

Train Dataset Example

<|im_start|>system
You are a helpful AI Assistant that provides well-reasoned and detailed responses. You first
think about the reasoning process as an internal monologue and then provide the user with
the answer. Respond in the following format: <think>
...
</think>
<answer>
...
</answer><|im_end|>
<|im_start|>user
You are a Couchbase SQL++ query expert. Given a database schema and a natural language question, generate a syntactically valid SQL++ query that precisely answers the question.

Rules:
- SELECT only the columns explicitly asked for — nothing more, nothing less
- Use the exact bucket, scope, and collection names provided in the database schema

Bucket Name:
`kaggle_bucket`

Scope Name:
`nudratabbas-sql-practice-dataset-3-hard-queries`

Database Schema:
{"`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`systems`": {"properties": {"criticality": {"samples": ["High", "Low"], "type": "string"}, "org_id": {"samples": ["ORG043", "ORG014"], "type": "string"}, "system_id": {"samples": ["S0059", "S0042"], "type": "string"}, "os_type": {"samples": ["Mac", "Linux"], "type": "string"}}, "type": "object"}, "`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`security_incidents`": {"properties": {"discovered_date": {"samples": ["2023-03-12", "2023-04-15"], "type": "string"}, "incident_type": {"samples": ["Data Breach", "Malware"], "type": "string"}, "incident_id": {"samples": ["I0056", "I0019"], "type": "string"}, "severity": {"samples": ["High", "Critical"], "type": "string"}, "org_id": {"samples": ["ORG009", "ORG017"], "type": "string"}}, "type": "object"}, "`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`users`": {"properties": {"org_id": {"samples": ["ORG029", "ORG006"], "type": "string"}, "role": {"samples": ["Analyst", "Admin"], "type": "string"}, "user_id": {"samples": ["U0168", "U0556"], "type": "string"}}, "type": "object"}, "`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`network_events`": {"properties": {"timestamp": {"samples": ["2023-03-14", "2023-04-30"], "type": "string"}, "event_type": {"samples": ["File Access", "Login"], "type": "string"}, "severity": {"samples": ["Critical", "High"], "type": "string"}, "system_id": {"samples": ["S0005", "S0107"], "type": "string"}, "event_id": {"samples": ["E01190", "E00188"], "type": "string"}}, "type": "object"}, "`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`login_logs`": {"properties": {"ip_address": {"samples": ["192.168.122.83", "192.168.119.166"], "type": "string"}, "status": {"samples": ["Success", "Failed"], "type": "string"}, "user_id": {"samples": ["U0097", "U0136"], "type": "string"}, "login_id": {"samples": ["L00204", "L02107"], "type": "string"}, "login_time": {"samples": ["2023-03-31", "2023-02-14"], "type": "string"}}, "type": "object"}, "`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`incident_systems`": {"properties": {"incident_id": {"samples": ["I0144", "I0035"], "type": "string"}, "system_id": {"samples": ["S0077", "S0012"], "type": "string"}}, "type": "object"}, "`kaggle_bucket`.`nudratabbas-sql-practice-dataset-3-hard-queries`.`organizations`": {"properties": {"industry": {"samples": ["Healthcare", "Finance"], "type": "string"}, "country": {"samples": ["Germany", "Canada"], "type": "string"}, "org_id": {"samples": ["ORG013", "ORG016"], "type": "string"}}, "type": "object"}}

This schema describes the structure of the data in the specified bucket and scope. It includes information about the collections, fields, and their data types.

Question:
Show organizations in Canada that had security incidents discovered after June 30, 2023.<|im_end|>
<|im_start|>assistant