Dataset Viewer
Auto-converted to Parquet Duplicate
code_to_key
unknown
codes
listlengths
2
18
criteria
unknown
fixture_id
stringlengths
8
35
gold_code
stringclasses
16 values
id
stringlengths
16
57
instructions
unknown
meta
dict
primitive
stringclasses
3 values
question_id
stringclasses
95 values
schema_version
stringclasses
1 value
split
stringclasses
1 value
state
unknown
teacher
unknown
teacher_model
stringclasses
1 value
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
choice-clear-billing
A
choice-clear-billing::department
"Which team should handle this ticket?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "routing" ] }
choice
department
oss-jev-distill-v1
train
"Help! My payouts have been failing for 3 days and I was charged twice."
{ "choice": "billing", "confidence": 1, "probabilities": { "billing": 1, "none_of_the_above": 0, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "refund", "B": "password_reset", "C": "none_of_the_above" }
[ "A", "B", "C" ]
{ "none_of_the_above": "Something else entirely", "password_reset": "Account access / password help", "refund": "Customer wants money back" }
choice-no-match-address
C
choice-no-match-address::intent
"What is the customer asking for?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "no-match" ] }
choice
intent
oss-jev-distill-v1
train
{ "message": "Please update my mailing address to 12 Oak St before Friday." }
{ "choice": "none_of_the_above", "confidence": 1, "probabilities": { "none_of_the_above": 1, "password_reset": 0, "refund": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2", "3": "3" }
[ "0", "1", "2", "3" ]
[ "completely unrelated", "adjacent field", "some direct experience", "deep, direct experience" ]
score-candidate-fit-pair
3
score-candidate-fit-pair::fit_a
"How relevant is `candidate_a` experience to `job`?"
{ "has_label": true, "suite_split": "train", "tags": [ "compare", "score" ] }
score
fit_a
oss-jev-distill-v1
train
{ "candidate_a": "Five years Django REST + PostgreSQL; owned production services.", "candidate_b": "Two years designing marketing landing pages in Webflow.", "job": "Senior backend engineer building Python APIs and PostgreSQL services." }
{ "confidence": 1, "legend": { "0": "completely unrelated", "1": "adjacent field", "2": "some direct experience", "3": "deep, direct experience" }, "probabilities": { "0": 0, "1": 0, "2": 0, "3": 1 }, "score": 3 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2", "3": "3" }
[ "0", "1", "2", "3" ]
[ "completely unrelated", "adjacent field", "some direct experience", "deep, direct experience" ]
score-candidate-fit-pair
0
score-candidate-fit-pair::fit_b
"How relevant is `candidate_b` experience to `job`?"
{ "has_label": true, "suite_split": "train", "tags": [ "compare", "score" ] }
score
fit_b
oss-jev-distill-v1
train
{ "candidate_a": "Five years Django REST + PostgreSQL; owned production services.", "candidate_b": "Two years designing marketing landing pages in Webflow.", "job": "Senior backend engineer building Python APIs and PostgreSQL services." }
{ "confidence": 0.93, "legend": { "0": "completely unrelated", "1": "adjacent field", "2": "some direct experience", "3": "deep, direct experience" }, "probabilities": { "0": 0.93, "1": 0.07, "2": 0, "3": 0 }, "score": 0.07 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "No urgency expressed", "true": "Explicitly time-sensitive or ASAP" }
noul-urgency-contrast
Y
noul-urgency-contrast::urgent_yes
"Does `urgent` convey time-sensitive urgency?"
{ "has_label": true, "suite_split": "train", "tags": [ "calibration", "noul" ] }
noul
urgent_yes
oss-jev-distill-v1
train
{ "chitchat": "Thanks, have a nice weekend!", "urgent": "My payouts failed for 3 days — please fix ASAP today." }
{ "noul": 0.99 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "No urgency expressed", "true": "Explicitly time-sensitive or ASAP" }
noul-urgency-contrast
N
noul-urgency-contrast::chitchat_urgent
"Does `chitchat` convey time-sensitive urgency?"
{ "has_label": true, "suite_split": "train", "tags": [ "calibration", "noul" ] }
noul
chitchat_urgent
oss-jev-distill-v1
train
{ "chitchat": "Thanks, have a nice weekend!", "urgent": "My payouts failed for 3 days — please fix ASAP today." }
{ "noul": 0.02 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for escalation", "true": "Explicitly asks for a person / urgent escalation" }
noul-soft-deadline
N
noul-soft-deadline::escalate
"Is the customer requesting immediate human escalation?"
{ "has_label": true, "suite_split": "train", "tags": [ "noul", "uncertain" ] }
noul
escalate
oss-jev-distill-v1
train
"Whenever you get a chance, could you look at invoice #4421? No rush if you're busy — just before month-end would be helpful."
{ "noul": 0.03 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales" }
[ "A", "B", "C" ]
{ "billing": "Payments, invoicing, refunds", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations" }
parallel-refund-ticket
A
parallel-refund-ticket::department
"Which team should handle `ticket` given `order`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score" ] }
choice
department
oss-jev-distill-v1
train
{ "order": { "charges": [ { "amount_usd": 49, "status": "captured" }, { "amount_usd": 49, "status": "captured" } ], "id": "A-104" }, "refund_policy": "Duplicate charges are eligible for a refund.", "ticket": { "messages": [ { ...
{ "choice": "billing", "confidence": 1, "probabilities": { "billing": 1, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Simple FAQ; scripted reply is enough", "Needs account lookup or policy check", "Escalation; ambiguous facts or high stakes" ]
parallel-refund-ticket
1
parallel-refund-ticket::complexity
"How complex is resolving this ticket?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score" ] }
score
complexity
oss-jev-distill-v1
train
{ "order": { "charges": [ { "amount_usd": 49, "status": "captured" }, { "amount_usd": 49, "status": "captured" } ], "id": "A-104" }, "refund_policy": "Duplicate charges are eligible for a refund.", "ticket": { "messages": [ { ...
{ "confidence": 0.8200000000000001, "legend": { "0": "Simple FAQ; scripted reply is enough", "1": "Needs account lookup or policy check", "2": "Escalation; ambiguous facts or high stakes" }, "probabilities": { "0": 0.11, "1": 0.88, "2": 0.01 }, "score": 0.9 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
parallel-refund-ticket
Y
parallel-refund-ticket::wants_refund
"Is the customer requesting a refund?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score" ] }
noul
wants_refund
oss-jev-distill-v1
train
{ "order": { "charges": [ { "amount_usd": 49, "status": "captured" }, { "amount_usd": 49, "status": "captured" } ], "id": "A-104" }, "refund_policy": "Duplicate charges are eligible for a refund.", "ticket": { "messages": [ { ...
{ "noul": 0.99 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
parallel-refund-ticket
Y
parallel-refund-ticket::policy_allows
"Does `refund_policy` support refunding this case?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score" ] }
noul
policy_allows
oss-jev-distill-v1
train
{ "order": { "charges": [ { "amount_usd": 49, "status": "captured" }, { "amount_usd": 49, "status": "captured" } ], "id": "A-104" }, "refund_policy": "Duplicate charges are eligible for a refund.", "ticket": { "messages": [ { ...
{ "noul": 0.98 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
parallel-refund-ticket
N
parallel-refund-ticket::is_technical_bug
"Is this primarily a software bug report?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score" ] }
noul
is_technical_bug
oss-jev-distill-v1
train
{ "order": { "charges": [ { "amount_usd": 49, "status": "captured" }, { "amount_usd": 49, "status": "captured" } ], "id": "A-104" }, "refund_policy": "Duplicate charges are eligible for a refund.", "ticket": { "messages": [ { ...
{ "noul": 0.13 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
state-string-billing
Y
state-string-billing::billing
"Is this about a billing issue?"
{ "has_label": true, "suite_split": "train", "tags": [ "noul", "state", "string" ] }
noul
billing
oss-jev-distill-v1
train
"My card was charged twice. Please refund."
{ "noul": 0.98 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales" }
[ "A", "B", "C" ]
{ "billing": "Payments and charges", "sales": "Pricing and upgrades", "technical": "Bugs, outages, API errors" }
state-nested-api-error
B
state-nested-api-error::dept
"Which team should handle `ticket.messages[0].text`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "nested", "state" ] }
choice
dept
oss-jev-distill-v1
train
{ "ticket": { "messages": [ { "from": "customer", "text": "The API returns 500 on /pay." }, { "from": "bot", "text": "Thanks, looking into it." } ] } }
{ "choice": "technical", "confidence": 0.9500000000000001, "probabilities": { "billing": 0.03, "sales": 0, "technical": 0.97 } }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
state-chat-array
Y
state-chat-array::duplicate_charge
"Does the conversation report a duplicate card charge?"
{ "has_label": true, "suite_split": "eval", "tags": [ "chat", "noul", "state" ] }
noul
duplicate_charge
oss-jev-distill-v1
train
[ { "content": "Hi", "role": "user" }, { "content": "My customer number is TS1337.", "role": "user" }, { "content": "My card was charged twice. Please fix.", "role": "user" } ]
{ "noul": 0.98 }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint" }
[ "A", "B", "C", "D" ]
{ "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
cookbook-route-return
C
cookbook-route-return::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "cookbook", "route" ] }
choice
intent
oss-jev-distill-v1
train
"I'd like to return the blue sweater from order A-104 for a refund."
{ "choice": "return_exchange", "confidence": 1, "probabilities": { "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 1 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Simple FAQ; scripted reply is enough", "Needs account lookup or policy check", "Escalation; ambiguous facts or high stakes" ]
cookbook-route-return
1
cookbook-route-return::complexity
"How complex is handling this request?"
{ "has_label": true, "suite_split": "eval", "tags": [ "cookbook", "route" ] }
score
complexity
oss-jev-distill-v1
train
"I'd like to return the blue sweater from order A-104 for a refund."
{ "confidence": 0.8200000000000001, "legend": { "0": "Simple FAQ; scripted reply is enough", "1": "Needs account lookup or policy check", "2": "Escalation; ambiguous facts or high stakes" }, "probabilities": { "0": 0.12, "1": 0.88, "2": 0 }, "score": 0.88 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
cookbook-rerank-refund
Y
cookbook-rerank-refund::rel_a
{ "passage": "Refunds for duplicate card charges are issued within 5 business days after verification.", "query": "How do I get a refund for a duplicate charge?", "question": "Would `passage` help answer `query`?" }
{ "has_label": true, "suite_split": "train", "tags": [ "cookbook", "noul", "rerank" ] }
noul
rel_a
oss-jev-distill-v1
train
{ "query": "How do I get a refund for a duplicate charge?" }
{ "noul": 0.63 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
cookbook-rerank-refund
N
cookbook-rerank-refund::rel_b
{ "passage": "Our warehouse ships most apparel orders within two business days.", "query": "How do I get a refund for a duplicate charge?", "question": "Would `passage` help answer `query`?" }
{ "has_label": true, "suite_split": "train", "tags": [ "cookbook", "noul", "rerank" ] }
noul
rel_b
oss-jev-distill-v1
train
{ "query": "How do I get a refund for a duplicate charge?" }
{ "noul": 0.01 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
null
cookbook-rerank-refund
N
cookbook-rerank-refund::rel_c
{ "passage": "To reset your password, open Settings and choose Security.", "query": "How do I get a refund for a duplicate charge?", "question": "Would `passage` help answer `query`?" }
{ "has_label": true, "suite_split": "train", "tags": [ "cookbook", "noul", "rerank" ] }
noul
rel_c
oss-jev-distill-v1
train
{ "query": "How do I get a refund for a duplicate charge?" }
{ "noul": 0.01 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-dup-charge
A
support-dup-charge::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I was charged twice for order A-104. Please refund the duplicate." }
{ "choice": "billing", "confidence": 1, "probabilities": { "billing": 1, "none_of_the_above": 0, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-dup-charge
C
support-dup-charge::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I was charged twice for order A-104. Please refund the duplicate." }
{ "choice": "return_exchange", "confidence": 1, "probabilities": { "account": 0, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 1 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-dup-charge
0
support-dup-charge::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I was charged twice for order A-104. Please refund the duplicate." }
{ "confidence": 0.93, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.9500000000000001, "1": 0.05, "2": 0 }, "score": 0.05 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-dup-charge
1
support-dup-charge::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I was charged twice for order A-104. Please refund the duplicate." }
{ "confidence": 0.37, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.02, "1": 0.58, "2": 0.4 }, "score": 1.38 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-dup-charge
N
support-dup-charge::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I was charged twice for order A-104. Please refund the duplicate." }
{ "noul": 0.04 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-api-500
B
support-api-500::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "POST /v1/pay returns HTTP 500 intermittently since Tuesday." }
{ "choice": "technical", "confidence": 0.84, "probabilities": { "billing": 0.12, "none_of_the_above": 0, "sales": 0, "technical": 0.88 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-api-500
D
support-api-500::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "POST /v1/pay returns HTTP 500 intermittently since Tuesday." }
{ "choice": "complaint", "confidence": 1, "probabilities": { "account": 0, "complaint": 1, "order_status": 0, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-api-500
0
support-api-500::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "POST /v1/pay returns HTTP 500 intermittently since Tuesday." }
{ "confidence": 0.9500000000000001, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.97, "1": 0.03, "2": 0 }, "score": 0.03 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-api-500
2
support-api-500::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "POST /v1/pay returns HTTP 500 intermittently since Tuesday." }
{ "confidence": 0.91, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0.06, "2": 0.9400000000000001 }, "score": 1.94 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-api-500
N
support-api-500::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "POST /v1/pay returns HTTP 500 intermittently since Tuesday." }
{ "noul": 0.04 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-lost-package
D
support-lost-package::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Tracking says delivered but nothing arrived for order #8821." }
{ "choice": "none_of_the_above", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 1, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-lost-package
A
support-lost-package::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Tracking says delivered but nothing arrived for order #8821." }
{ "choice": "order_status", "confidence": 0.9500000000000001, "probabilities": { "account": 0, "complaint": 0.04, "order_status": 0.96, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-lost-package
0
support-lost-package::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Tracking says delivered but nothing arrived for order #8821." }
{ "confidence": 0.30000000000000004, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.53, "1": 0.47000000000000003, "2": 0 }, "score": 0.47000000000000003 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-lost-package
2
support-lost-package::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Tracking says delivered but nothing arrived for order #8821." }
{ "confidence": 0.8200000000000001, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0.12, "2": 0.88 }, "score": 1.87 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-lost-package
N
support-lost-package::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Tracking says delivered but nothing arrived for order #8821." }
{ "noul": 0.03 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-wrong-size
D
support-wrong-size::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "The shoes arrived in size 9; I ordered 10. I'd like an exchange." }
{ "choice": "none_of_the_above", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 1, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-wrong-size
C
support-wrong-size::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "The shoes arrived in size 9; I ordered 10. I'd like an exchange." }
{ "choice": "return_exchange", "confidence": 1, "probabilities": { "account": 0, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 1 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-wrong-size
0
support-wrong-size::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "The shoes arrived in size 9; I ordered 10. I'd like an exchange." }
{ "confidence": 0.98, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.99, "1": 0.01, "2": 0 }, "score": 0.01 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-wrong-size
1
support-wrong-size::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "The shoes arrived in size 9; I ordered 10. I'd like an exchange." }
{ "confidence": 0.8, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.11, "1": 0.87, "2": 0.02 }, "score": 0.91 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-wrong-size
N
support-wrong-size::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "The shoes arrived in size 9; I ordered 10. I'd like an exchange." }
{ "noul": 0.03 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-password
B
support-password::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I can't log in after resetting my password three times." }
{ "choice": "technical", "confidence": 0.92, "probabilities": { "billing": 0, "none_of_the_above": 0.06, "sales": 0, "technical": 0.9400000000000001 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-password
E
support-password::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I can't log in after resetting my password three times." }
{ "choice": "account", "confidence": 1, "probabilities": { "account": 1, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-password
1
support-password::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I can't log in after resetting my password three times." }
{ "confidence": 0.8200000000000001, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.12, "1": 0.88, "2": 0 }, "score": 0.88 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-password
2
support-password::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I can't log in after resetting my password three times." }
{ "confidence": 0.9400000000000001, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0.04, "2": 0.96 }, "score": 1.96 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-password
N
support-password::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "I can't log in after resetting my password three times." }
{ "noul": 0.05 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-angry-outage
B
support-angry-outage::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Your app has been down all morning and I'm losing sales. Fix it NOW." }
{ "choice": "technical", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 0, "sales": 0, "technical": 1 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-angry-outage
D
support-angry-outage::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Your app has been down all morning and I'm losing sales. Fix it NOW." }
{ "choice": "complaint", "confidence": 1, "probabilities": { "account": 0, "complaint": 1, "order_status": 0, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-angry-outage
2
support-angry-outage::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Your app has been down all morning and I'm losing sales. Fix it NOW." }
{ "confidence": 0.76, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0, "1": 0.16, "2": 0.84 }, "score": 1.8399999999999999 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-angry-outage
2
support-angry-outage::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Your app has been down all morning and I'm losing sales. Fix it NOW." }
{ "confidence": 1, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0, "2": 1 }, "score": 2 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-angry-outage
N
support-angry-outage::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Your app has been down all morning and I'm losing sales. Fix it NOW." }
{ "noul": 0.06 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-polite-delay
A
support-polite-delay::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "No rush — whenever you can, please confirm my billing address update." }
{ "choice": "billing", "confidence": 0.97, "probabilities": { "billing": 0.98, "none_of_the_above": 0.02, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-polite-delay
E
support-polite-delay::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "No rush — whenever you can, please confirm my billing address update." }
{ "choice": "account", "confidence": 1, "probabilities": { "account": 1, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-polite-delay
0
support-polite-delay::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "No rush — whenever you can, please confirm my billing address update." }
{ "confidence": 1, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 1, "1": 0, "2": 0 }, "score": 0 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-polite-delay
0
support-polite-delay::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "No rush — whenever you can, please confirm my billing address update." }
{ "confidence": 0.9500000000000001, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.97, "1": 0.03, "2": 0 }, "score": 0.03 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-polite-delay
N
support-polite-delay::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "No rush — whenever you can, please confirm my billing address update." }
{ "noul": 0.04 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-invoice-q
A
support-invoice-q::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Can you explain line item 4 on invoice INV-2201?" }
{ "choice": "billing", "confidence": 1, "probabilities": { "billing": 1, "none_of_the_above": 0, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-invoice-q
B
support-invoice-q::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Can you explain line item 4 on invoice INV-2201?" }
{ "choice": "product_question", "confidence": 0.84, "probabilities": { "account": 0.1, "complaint": 0.02, "order_status": 0.01, "product_question": 0.87, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-invoice-q
0
support-invoice-q::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Can you explain line item 4 on invoice INV-2201?" }
{ "confidence": 1, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 1, "1": 0, "2": 0 }, "score": 0 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-invoice-q
1
support-invoice-q::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Can you explain line item 4 on invoice INV-2201?" }
{ "confidence": 0.46, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.35000000000000003, "1": 0.64, "2": 0.01 }, "score": 0.66 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-invoice-q
N
support-invoice-q::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Can you explain line item 4 on invoice INV-2201?" }
{ "noul": 0.07 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-webhook
B
support-webhook::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Our webhook endpoint stopped receiving payment.succeeded events." }
{ "choice": "technical", "confidence": 0.93, "probabilities": { "billing": 0.05, "none_of_the_above": 0, "sales": 0, "technical": 0.9500000000000001 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-webhook
D
support-webhook::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Our webhook endpoint stopped receiving payment.succeeded events." }
{ "choice": "complaint", "confidence": 0.91, "probabilities": { "account": 0.02, "complaint": 0.93, "order_status": 0.03, "product_question": 0.02, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-webhook
0
support-webhook::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Our webhook endpoint stopped receiving payment.succeeded events." }
{ "confidence": 0.9500000000000001, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.97, "1": 0.03, "2": 0 }, "score": 0.03 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-webhook
2
support-webhook::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Our webhook endpoint stopped receiving payment.succeeded events." }
{ "confidence": 0.97, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0.02, "2": 0.98 }, "score": 1.98 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-webhook
N
support-webhook::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Our webhook endpoint stopped receiving payment.succeeded events." }
{ "noul": 0.04 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-cancel-sub
A
support-cancel-sub::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Please cancel my subscription at the end of this billing period." }
{ "choice": "billing", "confidence": 0.96, "probabilities": { "billing": 0.97, "none_of_the_above": 0.02, "sales": 0.01, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-cancel-sub
E
support-cancel-sub::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Please cancel my subscription at the end of this billing period." }
{ "choice": "account", "confidence": 1, "probabilities": { "account": 1, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-cancel-sub
0
support-cancel-sub::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Please cancel my subscription at the end of this billing period." }
{ "confidence": 1, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 1, "1": 0, "2": 0 }, "score": 0 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-cancel-sub
1
support-cancel-sub::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Please cancel my subscription at the end of this billing period." }
{ "confidence": 0.39, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.36, "1": 0.59, "2": 0.05 }, "score": 0.6900000000000001 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-cancel-sub
N
support-cancel-sub::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Please cancel my subscription at the end of this billing period." }
{ "noul": 0.03 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-feature-how
D
support-feature-how::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "How do I export reports to CSV from the analytics dashboard?" }
{ "choice": "none_of_the_above", "confidence": 0.63, "probabilities": { "billing": 0, "none_of_the_above": 0.72, "sales": 0, "technical": 0.28 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-feature-how
B
support-feature-how::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "How do I export reports to CSV from the analytics dashboard?" }
{ "choice": "product_question", "confidence": 1, "probabilities": { "account": 0, "complaint": 0, "order_status": 0, "product_question": 1, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-feature-how
0
support-feature-how::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "How do I export reports to CSV from the analytics dashboard?" }
{ "confidence": 1, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 1, "1": 0, "2": 0 }, "score": 0 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-feature-how
0
support-feature-how::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "How do I export reports to CSV from the analytics dashboard?" }
{ "confidence": 0.27, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.51, "1": 0.49, "2": 0 }, "score": 0.5 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-feature-how
N
support-feature-how::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "How do I export reports to CSV from the analytics dashboard?" }
{ "noul": 0.02 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-partial-refund
A
support-partial-refund::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "One item in the bundle was damaged; refund just that SKU please." }
{ "choice": "billing", "confidence": 1, "probabilities": { "billing": 1, "none_of_the_above": 0, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-partial-refund
C
support-partial-refund::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "One item in the bundle was damaged; refund just that SKU please." }
{ "choice": "return_exchange", "confidence": 1, "probabilities": { "account": 0, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 1 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-partial-refund
0
support-partial-refund::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "One item in the bundle was damaged; refund just that SKU please." }
{ "confidence": 0.97, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.98, "1": 0.02, "2": 0 }, "score": 0.02 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-partial-refund
1
support-partial-refund::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "One item in the bundle was damaged; refund just that SKU please." }
{ "confidence": 0.8200000000000001, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.07, "1": 0.88, "2": 0.05 }, "score": 0.98 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-partial-refund
N
support-partial-refund::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "One item in the bundle was damaged; refund just that SKU please." }
{ "noul": 0.03 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-sso-fail
B
support-sso-fail::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Okta SSO login loops back to the IdP error page for our org." }
{ "choice": "technical", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 0, "sales": 0, "technical": 1 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-sso-fail
E
support-sso-fail::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Okta SSO login loops back to the IdP error page for our org." }
{ "choice": "account", "confidence": 1, "probabilities": { "account": 1, "complaint": 0, "order_status": 0, "product_question": 0, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-sso-fail
0
support-sso-fail::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Okta SSO login loops back to the IdP error page for our org." }
{ "confidence": 0.97, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.98, "1": 0.02, "2": 0 }, "score": 0.02 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-sso-fail
2
support-sso-fail::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Okta SSO login loops back to the IdP error page for our org." }
{ "confidence": 0.97, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0.02, "2": 0.98 }, "score": 1.98 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-sso-fail
N
support-sso-fail::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Okta SSO login loops back to the IdP error page for our org." }
{ "noul": 0.05 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-quote
C
support-quote::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We need a quote for 200 seats with SSO and audit logs." }
{ "choice": "sales", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 0, "sales": 1, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-quote
B
support-quote::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We need a quote for 200 seats with SSO and audit logs." }
{ "choice": "product_question", "confidence": 1, "probabilities": { "account": 0, "complaint": 0, "order_status": 0, "product_question": 1, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-quote
0
support-quote::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We need a quote for 200 seats with SSO and audit logs." }
{ "confidence": 1, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 1, "1": 0, "2": 0 }, "score": 0 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-quote
1
support-quote::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We need a quote for 200 seats with SSO and audit logs." }
{ "confidence": 0.61, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.25, "1": 0.74, "2": 0.01 }, "score": 0.75 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-quote
N
support-quote::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We need a quote for 200 seats with SSO and audit logs." }
{ "noul": 0.05 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-rate-limit
B
support-rate-limit::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We're hitting 429s on /search despite being on the Team plan." }
{ "choice": "technical", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 0, "sales": 0, "technical": 1 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-rate-limit
D
support-rate-limit::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We're hitting 429s on /search despite being on the Team plan." }
{ "choice": "complaint", "confidence": 0.96, "probabilities": { "account": 0.01, "complaint": 0.97, "order_status": 0, "product_question": 0.02, "return_exchange": 0 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-rate-limit
1
support-rate-limit::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We're hitting 429s on /search despite being on the Team plan." }
{ "confidence": 0.27, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 0.49, "1": 0.51, "2": 0 }, "score": 0.51 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-rate-limit
2
support-rate-limit::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We're hitting 429s on /search despite being on the Team plan." }
{ "confidence": 0.88, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0, "1": 0.08, "2": 0.92 }, "score": 1.92 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-rate-limit
N
support-rate-limit::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "eval", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "We're hitting 429s on /search despite being on the Team plan." }
{ "noul": 0.05 }
jev-1.13.0
{ "A": "billing", "B": "technical", "C": "sales", "D": "none_of_the_above" }
[ "A", "B", "C", "D" ]
{ "billing": "Payments, invoicing, refunds, payouts, charges", "none_of_the_above": "Does not fit billing, technical, or sales", "sales": "Pricing, upgrades, new accounts", "technical": "Bugs, outages, integrations, API errors" }
support-chatty
D
support-chatty::department
"Which team should handle `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
department
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Hi! Just saying thanks for the help last week. Have a great day." }
{ "choice": "none_of_the_above", "confidence": 1, "probabilities": { "billing": 0, "none_of_the_above": 1, "sales": 0, "technical": 0 } }
jev-1.13.0
{ "A": "order_status", "B": "product_question", "C": "return_exchange", "D": "complaint", "E": "account" }
[ "A", "B", "C", "D", "E" ]
{ "account": "Login, password, profile, or account settings", "complaint": "Angry complaint about service quality", "order_status": "Where is my order / tracking / shipment status", "product_question": "How does a product work / features / specs", "return_exchange": "Return, exchange, or refund a purchase" }
support-chatty
B
support-chatty::intent
"What is the customer's primary intent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
choice
intent
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Hi! Just saying thanks for the help last week. Have a great day." }
{ "choice": "product_question", "confidence": 0.36, "probabilities": { "account": 0.16, "complaint": 0.27, "order_status": 0.07, "product_question": 0.49, "return_exchange": 0.01 } }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Calm, just stating facts", "Frustrated but civil", "Very angry, strong language or threatening to leave" ]
support-chatty
0
support-chatty::frustration
"How frustrated is the customer in `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
frustration
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Hi! Just saying thanks for the help last week. Have a great day." }
{ "confidence": 1, "legend": { "0": "Calm, just stating facts", "1": "Frustrated but civil", "2": "Very angry, strong language or threatening to leave" }, "probabilities": { "0": 1, "1": 0, "2": 0 }, "score": 0 }
jev-1.13.0
{ "0": "0", "1": "1", "2": "2" }
[ "0", "1", "2" ]
[ "Can wait days or longer", "Should be handled this week", "Needs attention today / ASAP" ]
support-chatty
0
support-chatty::urgency
"How time-sensitive is resolving `ticket`?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
score
urgency
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Hi! Just saying thanks for the help last week. Have a great day." }
{ "confidence": 0.98, "legend": { "0": "Can wait days or longer", "1": "Should be handled this week", "2": "Needs attention today / ASAP" }, "probabilities": { "0": 0.99, "1": 0.01, "2": 0 }, "score": 0.01 }
jev-1.13.0
{ "N": "false", "Y": "true" }
[ "Y", "N" ]
{ "false": "Does not ask for a person", "true": "Explicitly asks for a person" }
support-chatty
N
support-chatty::wants_human
"Does `ticket` ask for a human agent?"
{ "has_label": true, "suite_split": "train", "tags": [ "choice", "noul", "parallel", "score", "support" ] }
noul
wants_human
oss-jev-distill-v1
train
{ "channel": "email", "ticket": "Hi! Just saying thanks for the help last week. Have a great day." }
{ "noul": 0.01 }
jev-1.13.0
End of preview. Expand in Data Studio

oss-jev-distill

Teacher-labeled distill pack for the one-token Qwen decision track (schema_version: oss-jev-distill-v1).

Counts (~10k pack)

Split Rows
train 7909
eval 1074
calib 1061
total 10044
fixtures labeled 1708

Teacher: jev-1.13.0.

Jobs mount

-v hf://datasets/gyanbhambhani/oss-jev-distill:/data

Root layout: train.jsonl, eval.jsonl, calib.jsonl, manifest.json. Do not train on eval or calib.

Downloads last month
243

Models trained or fine-tuned on gyanbhambhani/oss-jev-distill