Dataset Preview Go to dataset viewer
id (string)hit_ids (json)sentence (string)indices_into_review_text (json)model_0_label (string)model_0_probs (json)text_id (string)review_id (string)review_rating (int)label_distribution (json)gold_label (string)metadata (json)
r1-0000001
[ "y5238" ]
Roto-Rooter is always good when you need someone right away.
[ 0, 60 ]
positive
{ "negative": 0.01173639390617609, "positive": 0.7473671436309814, "neutral": 0.24089649319648743 }
r1-0000001
IDHkeGo-nxhqX4Exkdr08A
1
{ "positive": [ "w130", "w186", "w207", "w264", "w54" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000002
[ "y11155" ]
It's so worth the price of cox service over headaches of not being able to publish anything in a short amount of time on the Internet or frankly just stream in different rooms without buffering constantly.
[ 155, 360 ]
positive
{ "negative": 0.004590339958667755, "positive": 0.950273871421814, "neutral": 0.04513582959771156 }
r1-0000002
HERx4qHWIVDwH3AME9OYgQ
1
{ "positive": [ "w129", "w375", "w39" ], "negative": [ "w70" ], "neutral": [], "mixed": [ "w477" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000003
[ "y14984", "y17992" ]
I placed my order of "sticky ribs" as an appetizer and the "angry chicken" as my entree.
[ 530, 618 ]
negative
{ "negative": 0.9451407194137573, "positive": 0.004023305606096983, "neutral": 0.05083595588803291 }
r1-0000003
gCshZ3zPCaAwu8sAejmTcA
5
{ "positive": [ "w133" ], "negative": [], "neutral": [ "w135", "w184", "w38", "w439" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000004
[ "y1167", "y297" ]
There is mandatory valet parking, so make sure you get everything you need from the car!
[ 798, 886 ]
positive
{ "negative": 0.0416930615901947, "positive": 0.5428488254547119, "neutral": 0.4154581129550934 }
r1-0000004
JeLaSy9T3-pdVj_fQ6JX9A
5
{ "positive": [], "negative": [ "w135" ], "neutral": [ "w148", "w207", "w294", "w772" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000006
[ "y11412", "y19170", "y4073" ]
My wife and I couldn't finish it.
[ 294, 327 ]
negative
{ "negative": 0.977493166923523, "positive": 0.0006850706995464861, "neutral": 0.02182171680033207 }
r1-0000006
o-5hLEypDxR4jXjbmzkNBA
5
{ "positive": [], "negative": [ "w254" ], "neutral": [ "w176", "w328", "w39", "w598" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000007
[ "y2002" ]
I went with a revised quote and they set an appointment for the next week.
[ 732, 806 ]
negative
{ "negative": 0.6572147011756897, "positive": 0.09853064268827438, "neutral": 0.2442547082901001 }
r1-0000007
jttEKPQZ6UImUQVubG1GDg
5
{ "positive": [ "w280" ], "negative": [], "neutral": [ "w265", "w46", "w490", "w936" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000008
[ "y14885", "y16948" ]
I found out that the restaurant has only been open for about a week.
[ 138, 206 ]
negative
{ "negative": 0.45139288902282715, "positive": 0.12103912234306335, "neutral": 0.4275679886341095 }
r1-0000008
W8GvEfQ2qShtF7iRmehWVw
5
{ "positive": [], "negative": [], "neutral": [ "w137", "w310", "w415", "w460", "w627" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000010
[ "y18243", "y4069", "y4712" ]
I suffer with high blood pressure, headaches, type 2 diabetes and TMJ.
[ 449, 519 ]
neutral
{ "negative": 0.3392733931541443, "positive": 0.30037474632263184, "neutral": 0.36035192012786865 }
r1-0000010
uKrOYLnAVUsB1GE2huJt4A
5
{ "positive": [], "negative": [ "w155", "w193", "w204", "w620" ], "neutral": [ "w133" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000011
[ "y19430", "y2698" ]
Seriously, come here now before the hype sets in and you won't be able to get a table!
[ 1060, 1146 ]
positive
{ "negative": 0.026026399806141853, "positive": 0.8168579936027527, "neutral": 0.15711559355258942 }
r1-0000011
80r8He9zm4G1aR5N_7EP3Q
5
{ "positive": [ "w132", "w173", "w227" ], "negative": [], "neutral": [ "w447" ], "mixed": [ "w23" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000012
[ "y3416" ]
It was difficult to find the place.
[ 86, 121 ]
neutral
{ "negative": 0.2690688669681549, "positive": 0.2074701339006424, "neutral": 0.5234610438346863 }
r1-0000012
B6mQSD5qLmd0kaaTzRh8jw
5
{ "positive": [ "w385", "w591" ], "negative": [ "w290", "w539", "w552" ], "neutral": [], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000013
[ "y5784", "y801" ]
And their shuffleboard game.
[ 1282, 1310 ]
neutral
{ "negative": 0.041217926889657974, "positive": 0.04401402175426483, "neutral": 0.9147680997848511 }
r1-0000013
0DGlwRVW-m7RHGJOTWuR1A
5
{ "positive": [], "negative": [], "neutral": [ "w165", "w199", "w285", "w368", "w62" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000014
[ "y14474", "y19184" ]
Need a cheap spatula?
[ 542, 563 ]
neutral
{ "negative": 0.3583439886569977, "positive": 0.03304021805524826, "neutral": 0.6086158156394958 }
r1-0000014
aXrReHdGItNRnkijnVRH4g
5
{ "positive": [], "negative": [], "neutral": [ "w129", "w134", "w305", "w424", "w8" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000015
[ "y15969" ]
Mexican.
[ 7, 15 ]
neutral
{ "negative": 0.01613759435713291, "positive": 0.05673476681113243, "neutral": 0.9271275997161865 }
r1-0000015
FLqOVx0X5I4srqLMilNapQ
1
{ "positive": [], "negative": [], "neutral": [ "w143", "w188", "w625", "w707", "w716" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000016
[ "y11207", "y5687", "y7110" ]
Coronado Cafe and Kate's corner are still awesome but I think we will be venturing out to other casinos and see if we find out new spot.
[ 1520, 1656 ]
positive
{ "negative": 0.05092405527830124, "positive": 0.48395779728889465, "neutral": 0.4651181399822235 }
r1-0000016
pF1rD1WnVxbXddq3aRsAww
1
{ "positive": [ "w143", "w265", "w755" ], "negative": [], "neutral": [ "w577" ], "mixed": [ "w44" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000017
[ "y3946", "y8055" ]
Not sure why, just my observation.
[ 768, 802 ]
neutral
{ "negative": 0.0760495588183403, "positive": 0.0820464938879013, "neutral": 0.8419039249420166 }
r1-0000017
mh-RYGiocnxElg2BUltuyw
5
{ "positive": [], "negative": [], "neutral": [ "w198", "w210", "w22", "w35", "w671" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000019
[ "y1429" ]
You win!
[ 54, 62 ]
positive
{ "negative": 0.00790376216173172, "positive": 0.9436400532722473, "neutral": 0.048456232994794846 }
r1-0000019
nDW0K0d9rclQPLy5ZIgFAg
1
{ "positive": [ "w155", "w199", "w234", "w289" ], "negative": [], "neutral": [ "w326" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000020
[ "y11167", "y18763", "y580" ]
I'm not affiliated with Presto I was just super impressed by the food!
[ 1098, 1168 ]
positive
{ "negative": 0.0017010181909427047, "positive": 0.9760681390762329, "neutral": 0.02223076857626438 }
r1-0000020
kyzuLHLnr421QgJaNuGOxA
1
{ "positive": [ "w122", "w135", "w314", "w572" ], "negative": [], "neutral": [], "mixed": [ "w102" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000021
[ "y6427" ]
We told him again.
[ 3527, 3545 ]
neutral
{ "negative": 0.007084425538778305, "positive": 0.08201110363006592, "neutral": 0.9109045267105103 }
r1-0000021
rLA1DJqBZaigZsmC4FnlAw
1
{ "positive": [ "w187" ], "negative": [ "w321" ], "neutral": [ "w121", "w436", "w598" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000022
[ "y1837", "y9030" ]
Being that the place serves tapas-style plates, be prepared to spend a little more $$$.
[ 2288, 2375 ]
neutral
{ "negative": 0.17849618196487427, "positive": 0.12436950206756592, "neutral": 0.6971343159675598 }
r1-0000022
UbEhejiR-HxMMnHn6v9cEg
5
{ "positive": [], "negative": [ "w144", "w22", "w4" ], "neutral": [ "w296", "w57" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000023
[ "y12204", "y12390" ]
Don't kiss the desserts, not your typical.
[ 152, 194 ]
positive
{ "negative": 0.041412170976400375, "positive": 0.5366756916046143, "neutral": 0.42191219329833984 }
r1-0000023
BjN3IQJ43dojMdv9wl_Sdg
5
{ "positive": [], "negative": [ "w393" ], "neutral": [ "w129", "w153", "w222", "w60" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000024
[ "y17573", "y18578" ]
I went for dinner on a Sunday night.
[ 0, 36 ]
neutral
{ "negative": 0.31058475375175476, "positive": 0.19066913425922394, "neutral": 0.4987460970878601 }
r1-0000024
Gn99xywF2rb0cMhKM9q1YA
1
{ "positive": [], "negative": [], "neutral": [ "w135", "w212", "w439", "w832" ], "mixed": [ "w245" ] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000025
[ "y9164" ]
This is the place I always think of when a Strip hotel bar or swanky "lounge" charges me $12 for a precisely-measured (read: "stingy") something-and-soda.
[ 0, 154 ]
negative
{ "negative": 0.9054734110832214, "positive": 0.006846081931143999, "neutral": 0.08768061548471451 }
r1-0000025
4PUNdhRaBOkJM5GfJmADmA
5
{ "positive": [], "negative": [ "w16", "w193", "w444", "w566" ], "neutral": [], "mixed": [ "w299" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000026
[ "y11583" ]
Going to Arizona Tile!
[ 429, 451 ]
positive
{ "negative": 0.0066894968040287495, "positive": 0.9693523049354553, "neutral": 0.023958249017596245 }
r1-0000026
z3vCmcosGT_36M2hzdtnBQ
1
{ "positive": [ "w109", "w235" ], "negative": [], "neutral": [ "w114", "w188", "w50" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000028
[ "y15462", "y15471", "y15472" ]
How can the price go up in a recession!?!?!
[ 145, 188 ]
negative
{ "negative": 0.7989655137062073, "positive": 0.06025974452495575, "neutral": 0.14077478647232056 }
r1-0000028
hpTU0Ja222KsJx-CuvjIUA
5
{ "positive": [], "negative": [ "w239", "w279", "w32" ], "neutral": [ "w439" ], "mixed": [ "w123" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000029
[ "y14529", "y14827" ]
He deserves to be performing at a classier place.
[ 1354, 1403 ]
negative
{ "negative": 0.8333871960639954, "positive": 0.006685946602374315, "neutral": 0.15992677211761475 }
r1-0000029
bw3vosBf_rHAoS0eGdA3CA
5
{ "positive": [ "w160", "w205", "w58" ], "negative": [], "neutral": [ "w297", "w467" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000030
[ "y20182", "y7954" ]
We brought a bottle that we got from Napa.
[ 369, 411 ]
neutral
{ "negative": 0.3160594403743744, "positive": 0.06637421250343323, "neutral": 0.6175663471221924 }
r1-0000030
O-yk04uEf8eIFRmKHDZvvg
5
{ "positive": [], "negative": [], "neutral": [ "w247", "w33", "w415", "w544", "w85" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000031
[ "y13773", "y14834", "y20171" ]
We all ordered different things ranging from Ahi Tuna salad (I ordered it), harissa chicken, and tacos to pastas.
[ 500, 613 ]
neutral
{ "negative": 0.05926955118775368, "positive": 0.3318365514278412, "neutral": 0.608893871307373 }
r1-0000031
znhAL0FGtHI1QvfOWVaKcA
1
{ "positive": [], "negative": [], "neutral": [ "w135", "w195", "w32", "w395", "w40" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000032
[ "y8785", "y8786", "y8787" ]
At the desk helping another customer.
[ 2702, 2739 ]
neutral
{ "negative": 0.06068446487188339, "positive": 0.06318043917417526, "neutral": 0.8761351108551025 }
r1-0000032
5xg-_vPwlLn-FwhxSu5jpA
1
{ "positive": [], "negative": [], "neutral": [ "w140", "w176", "w311", "w55", "w858" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000033
[ "y20226", "y2623" ]
We got there for the 2eme service.
[ 272, 306 ]
negative
{ "negative": 0.5338059663772583, "positive": 0.07982441782951355, "neutral": 0.38636961579322815 }
r1-0000033
pVJeOZuF4LuZ_bcpyJUAEw
5
{ "positive": [], "negative": [], "neutral": [ "w132", "w224", "w26", "w39", "w609" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000034
[ "y16154", "y8881" ]
These are not coney dogs like Detroit Coney Grill down the street or James Coney Island hotdogs.
[ 244, 340 ]
negative
{ "negative": 0.6885123252868652, "positive": 0.05628788471221924, "neutral": 0.2551998198032379 }
r1-0000034
MfQp0XS26ReG1G0vg8n4VQ
1
{ "positive": [], "negative": [ "w1379", "w276" ], "neutral": [ "w161", "w50", "w57" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000035
[ "y3780" ]
It starts with the letter "T".
[ 1242, 1272 ]
neutral
{ "negative": 0.06260791420936584, "positive": 0.02929362840950489, "neutral": 0.9080985188484192 }
r1-0000035
9n4mMHpVvQUWQfFW7lYmLQ
5
{ "positive": [ "w536", "w630", "w641" ], "negative": [ "w931" ], "neutral": [], "mixed": [ "w599" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000037
[ "y14435", "y4772" ]
USED to love Otro.
[ 0, 18 ]
neutral
{ "negative": 0.3612070679664612, "positive": 0.019209854304790497, "neutral": 0.6195831298828125 }
r1-0000037
LQQa8_Te3WQ4TM1AYlFSmA
1
{ "positive": [], "negative": [ "w160", "w162", "w300" ], "neutral": [], "mixed": [ "w299", "w663" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000038
[ "y3622" ]
But thankfully 3 breads and rice.
[ 477, 510 ]
neutral
{ "negative": 0.20799744129180908, "positive": 0.11209027469158173, "neutral": 0.679912269115448 }
r1-0000038
O87cMw-B6Jp4Pvv5J6pH2Q
1
{ "positive": [ "w436", "w579", "w86" ], "negative": [ "w953" ], "neutral": [ "w374" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000039
[ "y2991" ]
We found some fabulous restaurants to explore.
[ 1357, 1403 ]
positive
{ "negative": 0.0007980158552527428, "positive": 0.9517825841903687, "neutral": 0.04741944000124931 }
r1-0000039
fcGZ-3FwAAh_WHYDZqhTwQ
1
{ "positive": [ "w135", "w296", "w539", "w871" ], "negative": [], "neutral": [], "mixed": [ "w970" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000041
[ "y14583", "y15728", "y903" ]
She said that would take too long but reluctantly went and got the stuff to do that.
[ 834, 918 ]
positive
{ "negative": 0.12145552784204483, "positive": 0.4516887366771698, "neutral": 0.4268557131290436 }
r1-0000041
CNcFth8s85jrIVEzcjBsVg
1
{ "positive": [], "negative": [ "w162", "w22", "w305" ], "neutral": [ "w297", "w55" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000042
[ "y14012" ]
If I had known this, I would have never switched.
[ 739, 788 ]
negative
{ "negative": 0.8083490133285522, "positive": 0.023676613345742226, "neutral": 0.1679743379354477 }
r1-0000042
-N2aUHvM2L1RHnyI5-cTCw
5
{ "positive": [], "negative": [ "w285", "w319", "w5", "w538" ], "neutral": [], "mixed": [ "w252" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000043
[ "y10860", "y4834" ]
I was very nervous to go to the DMV.
[ 0, 36 ]
negative
{ "negative": 0.5692477822303772, "positive": 0.17393523454666138, "neutral": 0.25681695342063904 }
r1-0000043
d1dxfYhDFLUDu4wjlRZZlg
5
{ "positive": [], "negative": [ "w126", "w143", "w223" ], "neutral": [ "w272" ], "mixed": [ "w177" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000044
[ "y10458", "y3697", "y4620" ]
But now I'm a confirmed kool aid drinker.
[ 198, 239 ]
neutral
{ "negative": 0.37934306263923645, "positive": 0.20504142343997955, "neutral": 0.4156154692173004 }
r1-0000044
EmaChzIl07oIkeWTu--Y1A
5
{ "positive": [ "w122", "w27", "w310" ], "negative": [], "neutral": [ "w343", "w4" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000045
[ "y10536" ]
I am a bit of a Barro's freak, and it's tops on my list for diet cheats...
[ 0, 74 ]
positive
{ "negative": 0.009173555299639702, "positive": 0.9459954500198364, "neutral": 0.044831033796072006 }
r1-0000045
RsV1kWoQaimlcVLiOoiEew
5
{ "positive": [ "w157", "w947", "w96" ], "negative": [ "w627" ], "neutral": [], "mixed": [ "w665" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000046
[ "y18639", "y8159" ]
Make sure you ask for Mike.
[ 330, 357 ]
positive
{ "negative": 0.004017828498035669, "positive": 0.5426563024520874, "neutral": 0.4533258378505707 }
r1-0000046
oT1iemk_FraA_LVZzQvTCQ
5
{ "positive": [ "w256", "w87" ], "negative": [], "neutral": [ "w26", "w328", "w447" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000047
[ "y13459", "y20210" ]
We figured we would have a few drinks and a burger... we ended up trying all the appetizers on their happy hour menu and we tried their surf and turf burger.
[ 241, 398 ]
neutral
{ "negative": 0.03101534955203533, "positive": 0.4221387505531311, "neutral": 0.5468458533287048 }
r1-0000047
KanJQZQweORKCmz9XQcT8A
5
{ "positive": [ "w23", "w54", "w620" ], "negative": [], "neutral": [ "w35", "w542" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000048
[ "y6297" ]
Always in a hurry in the morning.
[ 49, 82 ]
neutral
{ "negative": 0.2315583974123001, "positive": 0.0809391587972641, "neutral": 0.687502384185791 }
r1-0000048
y5QpRtWPVqPXEbekHi3WSQ
5
{ "positive": [ "w666" ], "negative": [], "neutral": [ "w160", "w517", "w55", "w76" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000049
[ "y16365", "y5303", "y7031" ]
The whole family was excited to know there was a new restaurant near by.
[ 26, 98 ]
positive
{ "negative": 0.0032109906896948814, "positive": 0.9437807202339172, "neutral": 0.05300825089216232 }
r1-0000049
UrJqz9NOmhGM_rKfUE1vzw
1
{ "positive": [ "w157", "w221", "w55" ], "negative": [], "neutral": [ "w26", "w43" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000050
[ "y20509", "y8279" ]
nor will you like 42% of restaurants on yelp.
[ 655, 700 ]
negative
{ "negative": 0.7531530261039734, "positive": 0.009166456758975983, "neutral": 0.23768050968647003 }
r1-0000050
_csyE3c9HSiz5OWs8L_Y5A
5
{ "positive": [], "negative": [], "neutral": [ "w130", "w153", "w240", "w4", "w45" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000051
[ "y13037", "y17918" ]
I never actually met this promoter.
[ 292, 327 ]
neutral
{ "negative": 0.34164538979530334, "positive": 0.08946441859006882, "neutral": 0.5688900947570801 }
r1-0000051
RX68QBYc8mV7qoRyhy09dA
5
{ "positive": [], "negative": [], "neutral": [ "w1131", "w265", "w516", "w734", "w84" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000052
[ "y1577" ]
How can he miss us???
[ 665, 686 ]
negative
{ "negative": 0.699152946472168, "positive": 0.09727346152067184, "neutral": 0.203573539853096 }
r1-0000052
_NVeHPnFXeX6PpdAg6TUmw
1
{ "positive": [ "w201" ], "negative": [ "w128", "w165", "w318" ], "neutral": [ "w545" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000053
[ "y11835", "y19259" ]
Nothing comes even close to being half as good on quality, service, selection, pricing, knowledge, financing, etc.
[ 703, 817 ]
negative
{ "negative": 0.9755499958992004, "positive": 0.007948880083858967, "neutral": 0.016501160338521004 }
r1-0000053
2jKuxJgeg4m7r5p3DgoX1w
5
{ "positive": [ "w100", "w294", "w52", "w527", "w889" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000054
[ "y16231" ]
STEER CLEAR!
[ 0, 12 ]
positive
{ "negative": 0.07518152892589569, "positive": 0.46613067388534546, "neutral": 0.45868781208992004 }
r1-0000054
_M3HUJkWvyV0I1uNlSyxjA
1
{ "positive": [ "w264" ], "negative": [ "w166", "w23", "w3", "w401" ], "neutral": [], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000055
[ "y2339" ]
Loud bang.
[ 802, 812 ]
neutral
{ "negative": 0.0176137313246727, "positive": 0.4690435826778412, "neutral": 0.5133426785469055 }
r1-0000055
rXUL2N9WuLlnvGdPeCsecw
1
{ "positive": [], "negative": [ "w17", "w990" ], "neutral": [ "w386", "w415", "w49" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000056
[ "y16412" ]
Its a safeway company so if you are from Cali use your Safeway Club Card.
[ 57, 130 ]
neutral
{ "negative": 0.19671232998371124, "positive": 0.3089805841445923, "neutral": 0.4943070411682129 }
r1-0000056
XEetuhHYDhMUUuZetX8uGQ
5
{ "positive": [ "w385", "w685" ], "negative": [], "neutral": [ "w130", "w52", "w840" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000057
[ "y17219", "y18027", "y4371" ]
The spinach was pretty flavorless, though.
[ 1034, 1076 ]
negative
{ "negative": 0.6507993936538696, "positive": 0.01997697725892067, "neutral": 0.3292236626148224 }
r1-0000057
kDKKtaUrfMbNbA9CuDhwpg
5
{ "positive": [], "negative": [ "w26", "w279", "w408", "w79", "w862" ], "neutral": [], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000058
[ "y16136", "y7841" ]
I booked this hotel for a room on Jan 2 2015.
[ 66, 111 ]
negative
{ "negative": 0.9394829869270325, "positive": 0.015047682449221611, "neutral": 0.04546929523348808 }
r1-0000058
FZryCvEUrtPZBFJxym1XFg
1
{ "positive": [], "negative": [], "neutral": [ "w100", "w231", "w26", "w279", "w425" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000059
[ "y10361", "y12272", "y9827" ]
We ended up here Saturday afternoon as we waited for the ASU tailgating scene to get going (which it never really did).
[ 0, 119 ]
negative
{ "negative": 0.5339961647987366, "positive": 0.06769336014986038, "neutral": 0.39831045269966125 }
r1-0000059
QUMr-uD10PptKRi-TOZUsQ
5
{ "positive": [], "negative": [ "w193", "w36", "w760" ], "neutral": [ "w17", "w48" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000060
[ "y1892", "y3074" ]
Dog friendly?
[ 0, 13 ]
neutral
{ "negative": 0.14081043004989624, "positive": 0.22970275580883026, "neutral": 0.6294868588447571 }
r1-0000060
zfwmtqh44cMvTq0D-dxXhg
1
{ "positive": [ "w119", "w388" ], "negative": [], "neutral": [ "w211", "w26", "w290" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000061
[ "y4513", "y4632" ]
After measuring it we figured it would fit in our trunk but it didn't.
[ 115, 185 ]
negative
{ "negative": 0.5146036148071289, "positive": 0.061822276562452316, "neutral": 0.42357414960861206 }
r1-0000061
58xPo_OHDFLDQF5h5o5JRg
5
{ "positive": [], "negative": [ "w155", "w161", "w188" ], "neutral": [ "w8" ], "mixed": [ "w457" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000062
[ "y4735" ]
I have eaten at this place many times over the years and always been happy with the food and the service.
[ 0, 105 ]
positive
{ "negative": 0.005712695419788361, "positive": 0.9148672223091125, "neutral": 0.07942002266645432 }
r1-0000062
PN03CRT5CUmsS1frIP998A
1
{ "positive": [ "w226", "w311", "w35", "w463", "w54" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000063
[ "y2228", "y2750", "y8975" ]
is unbarable.
[ 246, 259 ]
positive
{ "negative": 0.004676783457398415, "positive": 0.88719642162323, "neutral": 0.10812681168317795 }
r1-0000063
LmjXv4ooD3qSvodSs8F2Aw
1
{ "positive": [ "w342", "w474" ], "negative": [ "w288", "w352", "w447" ], "neutral": [], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000064
[ "y15403", "y16180" ]
I call again.
[ 1139, 1152 ]
neutral
{ "negative": 0.012055614031851292, "positive": 0.08838760852813721, "neutral": 0.8995567560195923 }
r1-0000064
eL7ZinuyMmW27uq_Lraz2Q
1
{ "positive": [], "negative": [ "w183" ], "neutral": [ "w122", "w1748", "w34", "w755" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000065
[ "y20186", "y7804", "y9147" ]
We asked for sriracha sauce three times to three different people before finally giving up.
[ 389, 480 ]
negative
{ "negative": 0.9765224456787109, "positive": 0.0004989303415641189, "neutral": 0.022978641092777252 }
r1-0000065
DAH-sFwceKg5vRjc3iGNPw
5
{ "positive": [ "w23" ], "negative": [ "w17", "w222", "w365", "w41" ], "neutral": [], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000066
[ "y17829", "y8360" ]
I was seated and they explained how everything worked...
[ 72, 129 ]
positive
{ "negative": 0.023700831457972527, "positive": 0.7174158692359924, "neutral": 0.2588832974433899 }
r1-0000066
cwBBMP-TUyfodX23LKXOEA
1
{ "positive": [ "w223", "w258" ], "negative": [], "neutral": [ "w408", "w57", "w609" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000068
[ "y4233" ]
Not the case with Irrigation Repair Service of Gilbert (IRS - don't be afraid of the acronym), which supplied the utmost competence on all three selling points.
[ 174, 334 ]
positive
{ "negative": 0.1365562230348587, "positive": 0.8118839263916016, "neutral": 0.051559895277023315 }
r1-0000068
TUsHOJo4B3nKJ3TWH-gqsw
5
{ "positive": [ "w114", "w128", "w226", "w376", "w57" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000070
[ "y12192", "y1297" ]
Do not buy from them until you compare!!!!
[ 238, 280 ]
negative
{ "negative": 0.9605700373649597, "positive": 0.012229292653501034, "neutral": 0.027200721204280853 }
r1-0000070
sMm3UGAmpF6W64MfBTU_JA
5
{ "positive": [], "negative": [ "w100", "w420", "w490", "w617" ], "neutral": [ "w80" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000071
[ "y10423" ]
It's served alongside our garlic bread and made fresh daily, so get yours before we run out!"
[ 386, 479 ]
positive
{ "negative": 0.0009642356890253723, "positive": 0.9684404730796814, "neutral": 0.030595267191529274 }
r1-0000071
kUZCmocgt0UrsbznGYY33Q
1
{ "positive": [ "w186", "w333", "w432", "w471", "w8" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000073
[ "y1534", "y2821" ]
really old guy missing teeth.
[ 100, 129 ]
neutral
{ "negative": 0.12784337997436523, "positive": 0.029340891167521477, "neutral": 0.8428156971931458 }
r1-0000073
i8jZfuGAeUF81aUKYkr93g
5
{ "positive": [], "negative": [ "w165", "w212", "w250", "w639" ], "neutral": [], "mixed": [ "w341" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000074
[ "y11199", "y9286" ]
I wasn't even issue a new key card when the gym transitioned from Gold's to EOS.
[ 467, 547 ]
negative
{ "negative": 0.822490930557251, "positive": 0.06771635264158249, "neutral": 0.10979270190000534 }
r1-0000074
QQi-9N2-aK3iyjAK3vmUNw
1
{ "positive": [], "negative": [ "w108", "w395", "w8" ], "neutral": [ "w290", "w71" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000075
[ "y5908", "y6407", "y6423" ]
An update on Peoria Ford.
[ 0, 25 ]
neutral
{ "negative": 0.07865799218416214, "positive": 0.41396957635879517, "neutral": 0.5073723793029785 }
r1-0000075
L4qc79vlYBaOBjaSjUFLVw
1
{ "positive": [], "negative": [], "neutral": [ "w143", "w32", "w380", "w80", "w832" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000076
[ "y995" ]
He did a decent job on my fixes.
[ 414, 446 ]
positive
{ "negative": 0.0271324273198843, "positive": 0.5014843940734863, "neutral": 0.4713832139968872 }
r1-0000076
od2_4WZlcXM3PlZ7ElYS9w
1
{ "positive": [ "w417", "w630", "w727" ], "negative": [], "neutral": [ "w369" ], "mixed": [ "w76" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000077
[ "y11045", "y420" ]
Can't wait til the next time.
[ 537, 566 ]
positive
{ "negative": 0.013300797902047634, "positive": 0.9646247029304504, "neutral": 0.022074466571211815 }
r1-0000077
5gYJQekxyOe-1-H7R27KMQ
5
{ "positive": [ "w135", "w143", "w299", "w482" ], "negative": [], "neutral": [ "w23" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000078
[ "y11747", "y19348" ]
Our server, Jesse, took our orders.
[ 218, 253 ]
positive
{ "negative": 0.010599385015666485, "positive": 0.764857828617096, "neutral": 0.2245427817106247 }
r1-0000078
F_FoyRf7srm2dlNgw2PFGg
1
{ "positive": [], "negative": [], "neutral": [ "w101", "w162", "w180", "w305", "w311" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000080
[ "y16498", "y5279" ]
I didn't just remember those things and put them in my own run on sentence, that is how she said it.
[ 1767, 1867 ]
neutral
{ "negative": 0.37396419048309326, "positive": 0.19372722506523132, "neutral": 0.4323086142539978 }
r1-0000080
mPOdSV56QdsoZ57vgAdvDg
1
{ "positive": [], "negative": [ "w276" ], "neutral": [ "w140", "w27", "w299", "w396" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000081
[ "y18371", "y18554", "y7903" ]
Mine was an older shuffle iPod and hubby's was a newly purchased iPod touch, wasn't holding a charge and kept dying, as the battery kept draining ever so quickly (still under warranty, until next June).
[ 583, 785 ]
positive
{ "negative": 0.017564402893185616, "positive": 0.9408647418022156, "neutral": 0.04157085716724396 }
r1-0000081
ZcBUz52uSyE0Pa6wyy89kQ
5
{ "positive": [], "negative": [ "w143", "w305", "w34" ], "neutral": [ "w68" ], "mixed": [ "w837" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000082
[ "y19323", "y3738", "y3739" ]
Only been here once.
[ 0, 20 ]
neutral
{ "negative": 0.23316945135593414, "positive": 0.04900443181395531, "neutral": 0.7178261280059814 }
r1-0000082
Z0i9PR2eFxVFumuFNt_xNg
5
{ "positive": [], "negative": [], "neutral": [ "w1254", "w37", "w380", "w55", "w79" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000084
[ "y3230", "y3231", "y507" ]
Our bill was over a grand after everything was said and done.
[ 1540, 1601 ]
negative
{ "negative": 0.7608275413513184, "positive": 0.0384473092854023, "neutral": 0.20072518289089203 }
r1-0000084
pfo5P8u7kG9v0Zt1HYnifQ
5
{ "positive": [ "w641" ], "negative": [], "neutral": [ "w17", "w19", "w360", "w698" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000085
[ "y5772", "y5920" ]
Also, I am not Korean and I am usually one of a few Caucasians in the restaurant when I am here.
[ 482, 578 ]
neutral
{ "negative": 0.46218833327293396, "positive": 0.0746566653251648, "neutral": 0.46315500140190125 }
r1-0000085
HHtYoctYsBqDUcVgMjscOg
5
{ "positive": [ "w1305", "w72" ], "negative": [], "neutral": [ "w135", "w163", "w382" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000086
[ "y16270", "y17874" ]
I was asked the obligatory question that every person that ever worked retail is to ask a customer.
[ 469, 568 ]
neutral
{ "negative": 0.04620305076241493, "positive": 0.15265889465808868, "neutral": 0.8011381030082703 }
r1-0000086
Qp4X7baY_7StqgPHRWbttw
5
{ "positive": [], "negative": [ "w135" ], "neutral": [ "w122", "w163", "w295", "w350" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000087
[ "y15957", "y8020" ]
They have unseated our defacto Thai restaurant and will now become of our our regular spots.
[ 295, 387 ]
negative
{ "negative": 0.9566402435302734, "positive": 0.03315422683954239, "neutral": 0.010205572471022606 }
r1-0000087
v-oI-M2tujtkGClMn5GStA
5
{ "positive": [ "w229", "w240" ], "negative": [], "neutral": [ "w149", "w157", "w23" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000088
[ "y1712", "y19198" ]
No liquor license and not sure if you can byob (cold Pacifico would have been nice).
[ 421, 505 ]
negative
{ "negative": 0.4608093798160553, "positive": 0.09178459644317627, "neutral": 0.4474060535430908 }
r1-0000088
Tw30QZWN9Tw1yenOCqxNAQ
5
{ "positive": [], "negative": [ "w157", "w22" ], "neutral": [ "w106", "w128", "w279" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000089
[ "y16624", "y1955" ]
I'm glad they could get rid of those bees.
[ 911, 953 ]
positive
{ "negative": 0.05263874679803848, "positive": 0.6592355370521545, "neutral": 0.2881257236003876 }
r1-0000089
pMTjEpeZafQZnZp2Z8XKnw
5
{ "positive": [ "w1256", "w52", "w632", "w869" ], "negative": [], "neutral": [ "w45" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000090
[ "y20418", "y7954" ]
With Nellis AFB in the area, you would think there would be more surplus stores, but there are not.
[ 1077, 1176 ]
neutral
{ "negative": 0.32454076409339905, "positive": 0.08172517269849777, "neutral": 0.593734085559845 }
r1-0000090
EI25c_8M1KhhGZEOnM62MA
5
{ "positive": [], "negative": [ "w213", "w247" ], "neutral": [ "w415", "w447", "w544" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000091
[ "y15557", "y9170" ]
If someone isn't ready when they say they are, don't wait for them because folks who are ready to go when they say they are expect you to be!
[ 543, 684 ]
positive
{ "negative": 0.060660071671009064, "positive": 0.8996450304985046, "neutral": 0.03969492018222809 }
r1-0000091
spnjcNHgDnT2f6siNnkAuQ
1
{ "positive": [ "w179", "w553", "w568" ], "negative": [ "w17" ], "neutral": [ "w390" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000093
[ "y11599", "y3104", "y7086" ]
Its a take-out kind of place.
[ 34, 63 ]
neutral
{ "negative": 0.2126961648464203, "positive": 0.03711751848459244, "neutral": 0.7501863241195679 }
r1-0000093
Q1tQEcs8VuBLMTYubbOQ9Q
5
{ "positive": [ "w258", "w489" ], "negative": [], "neutral": [ "w299", "w71", "w737" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000094
[ "y11765" ]
The third transaction went smooth.
[ 626, 660 ]
positive
{ "negative": 0.002713320776820183, "positive": 0.9466528296470642, "neutral": 0.05063378065824509 }
r1-0000094
jqGT4X-0bOgK4pcBfejA3Q
1
{ "positive": [ "w515", "w725", "w790", "w842" ], "negative": [], "neutral": [ "w1347" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000095
[ "y12208", "y236", "y4826" ]
Do note that it IS supposed to be more bitter than sweet.
[ 31, 88 ]
neutral
{ "negative": 0.3217654824256897, "positive": 0.13603706657886505, "neutral": 0.5421974658966064 }
r1-0000095
s3XO_CwWleezivnDDyX8ow
5
{ "positive": [ "w23" ], "negative": [], "neutral": [ "w620", "w68", "w71" ], "mixed": [ "w223" ] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000096
[ "y450" ]
The white and red sangrias were lovely.
[ 928, 967 ]
positive
{ "negative": 0.006733091082423925, "positive": 0.7922220826148987, "neutral": 0.20104487240314484 }
r1-0000096
viJEF0iTv6NshyEptiuHRg
1
{ "positive": [ "w116", "w184", "w375", "w464", "w87" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000097
[ "y7193" ]
My car way sooo bad.
[ 54, 74 ]
negative
{ "negative": 0.9249998331069946, "positive": 0.024158049374818802, "neutral": 0.05084206908941269 }
r1-0000097
hlitKEmQ29DI1DDjbDCY1g
5
{ "positive": [], "negative": [ "w1021", "w135", "w152", "w201", "w491" ], "neutral": [], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000098
[ "y10015", "y18269" ]
The bathroom was clean.
[ 437, 460 ]
neutral
{ "negative": 0.03929327800869942, "positive": 0.354475200176239, "neutral": 0.6062314510345459 }
r1-0000098
lhA6eGPE2H3N-Xa5FdUtyQ
5
{ "positive": [ "w184", "w370", "w778", "w945" ], "negative": [], "neutral": [ "w646" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000099
[ "y12409", "y12450" ]
Don't waste your time- give them a call!
[ 395, 435 ]
negative
{ "negative": 0.9967790246009827, "positive": 0.000891755276825279, "neutral": 0.0023292251862585545 }
r1-0000099
8-HYEnxljk3mUHuoHUtU6Q
5
{ "positive": [ "w288", "w52", "w613" ], "negative": [ "w1630" ], "neutral": [ "w447" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000100
[ "y13713", "y16904", "y2426" ]
The waitress even setup her own mobile hotspot for me to use as the restaurant's wi-fi was not working.
[ 731, 834 ]
positive
{ "negative": 0.03375811502337456, "positive": 0.6903913617134094, "neutral": 0.2758505046367645 }
r1-0000100
boie3B2gnvVXSxoWQfrM9Q
5
{ "positive": [ "w111", "w1367", "w17", "w178", "w55" ], "negative": [], "neutral": [], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000101
[ "y2367" ]
I figured I had to go to this specific department because the sweaters were found here.
[ 567, 654 ]
neutral
{ "negative": 0.08521131426095963, "positive": 0.3711188733577728, "neutral": 0.5436697602272034 }
r1-0000101
DdrD1aThrjHGtTl20JsMEw
1
{ "positive": [ "w1066", "w14" ], "negative": [], "neutral": [ "w448", "w577", "w81" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000102
[ "y18158", "y2140", "y7196" ]
The crostini looked like a cheesy bread but ended up tasting more like a cheesy bread on steroids.
[ 1810, 1908 ]
negative
{ "negative": 0.9607546329498291, "positive": 0.0007049888372421265, "neutral": 0.03854048252105713 }
r1-0000102
macUlFgt9KW81mIdMrO1Vg
5
{ "positive": [ "w1786", "w216", "w333" ], "negative": [ "w386" ], "neutral": [], "mixed": [ "w135" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000103
[ "y3621" ]
I went to the pool to party and had the best 4 hour nap.
[ 23, 79 ]
positive
{ "negative": 0.028758781030774117, "positive": 0.938450276851654, "neutral": 0.032790981233119965 }
r1-0000103
_4EYIujSzAQsz5HCQTTSUg
5
{ "positive": [ "w201", "w30", "w363", "w474" ], "negative": [], "neutral": [], "mixed": [ "w805" ] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000104
[ "y2398" ]
I don't know what is going on at Stephanie street in Henderson, but it is turning into restaurant and junk food rowe of the southeast!
[ 1528, 1662 ]
negative
{ "negative": 0.5198332071304321, "positive": 0.2143346071243286, "neutral": 0.26583218574523926 }
r1-0000104
sHRFZIkl8TNE1Zva8u_36Q
5
{ "positive": [], "negative": [], "neutral": [ "w328", "w36", "w4" ], "mixed": [ "w666", "w972" ] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000105
[ "y1285" ]
Books trickle out of crevices of walls and shelf's alike.
[ 267, 324 ]
neutral
{ "negative": 0.11211304366588593, "positive": 0.17811284959316254, "neutral": 0.7097740769386292 }
r1-0000105
HHpRh50h8OkPo6XroWDH4Q
5
{ "positive": [], "negative": [], "neutral": [ "w130", "w139", "w241", "w461", "w84" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000106
[ "y9203" ]
When you place your order, they will give you a number to place on your table for the food to be delivered.
[ 164, 271 ]
positive
{ "negative": 0.23111717402935028, "positive": 0.4442256987094879, "neutral": 0.324657142162323 }
r1-0000106
bGXehDzmhMNEsi1UeFJGjQ
5
{ "positive": [ "w109" ], "negative": [], "neutral": [ "w129", "w1490", "w155", "w252" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000107
[ "y4074" ]
I was curious about the Seafood Risotto for a kosher place, are they going to use Pollock?
[ 1324, 1414 ]
neutral
{ "negative": 0.11829402297735214, "positive": 0.3616541624069214, "neutral": 0.5200517773628235 }
r1-0000107
FwtqHxxqrvIRTK4l_g_m1g
1
{ "positive": [], "negative": [ "w938" ], "neutral": [ "w132", "w148", "w26", "w59" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000108
[ "y13573" ]
I ordered to go so I could make it to my appointment on time.
[ 189, 250 ]
positive
{ "negative": 0.07309255003929138, "positive": 0.5847066640853882, "neutral": 0.34220078587532043 }
r1-0000108
M0G5HEtRGjP3GwqJzH4xXw
1
{ "positive": [ "w27" ], "negative": [], "neutral": [ "w16", "w263", "w305", "w8" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000109
[ "y11974", "y1660" ]
Definitely too much for one person.
[ 281, 316 ]
neutral
{ "negative": 0.4212559163570404, "positive": 0.08036341518163681, "neutral": 0.4983806312084198 }
r1-0000109
Z664_SVwOxTusiLqkKKDuw
5
{ "positive": [], "negative": [ "w258", "w57", "w879" ], "neutral": [ "w195", "w490" ], "mixed": [] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000110
[ "y13636", "y6360" ]
Since I started seeing Dr. Mayer, the wait time post check in has always been less than 5 minutes for the pre-screening workout and then anither 5-10 minutes (at most) to see the doctor after being placed in an exam room.
[ 252, 473 ]
positive
{ "negative": 0.21935796737670898, "positive": 0.5343391299247742, "neutral": 0.24630293250083923 }
r1-0000110
kMlRVNP2kyksPr4uQ2LlUg
5
{ "positive": [ "w18", "w211", "w53", "w54" ], "negative": [], "neutral": [ "w26" ], "mixed": [] }
positive
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000111
[ "y13490" ]
First saw a Hakkasan in San Francisco but they were booked for the evening so we never got in.
[ 0, 94 ]
neutral
{ "negative": 0.1411745548248291, "positive": 0.363070011138916, "neutral": 0.4957554042339325 }
r1-0000111
IUeVcPo-qp0jfth0cG50Eg
5
{ "positive": [ "w521" ], "negative": [ "w115", "w143", "w22" ], "neutral": [], "mixed": [ "w525" ] }
negative
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
r1-0000112
[ "y4237", "y8266" ]
My dogs never barked, so even if I missed the knocking, they wouldn't have.
[ 500, 575 ]
positive
{ "negative": 0.014877786859869957, "positive": 0.8939515352249146, "neutral": 0.09117067605257034 }
r1-0000112
qZmm5Kxp4e3qyPBjPLoD1A
1
{ "positive": [], "negative": [ "w43" ], "neutral": [ "w288", "w326", "w48", "w594" ], "mixed": [] }
neutral
{ "split": "train", "round": 1, "subset": "all", "model_in_the_loop": "RoBERTa" }
End of preview (truncated to 100 rows)
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/datasets-cards)

DynaSent: Dynamic Sentiment Analysis Dataset

DynaSent is an English-language benchmark task for ternary (positive/negative/neutral) sentiment analysis. This dataset card is forked from the original DynaSent Repository.

Contents

Citation

Christopher Potts, Zhengxuan Wu, Atticus Geiger, and Douwe Kiela. 2020. DynaSent: A dynamic benchmark for sentiment analysis. Ms., Stanford University and Facebook AI Research.

  @article{potts-etal-2020-dynasent,
    title={{DynaSent}: A Dynamic Benchmark for Sentiment Analysis},
    author={Potts, Christopher and Wu, Zhengxuan and Geiger, Atticus and Kiela, Douwe},
    journal={arXiv preprint arXiv:2012.15349},
    url={https://arxiv.org/abs/2012.15349},
    year={2020}}

Dataset files

The dataset is dynasent-v1.1.zip, which is included in this repository. v1.1 differs from v1 only in that v1.1 has proper unique ids for Round 1 and corrects a bug that led to some non-unique ids in Round 2. There are no changes to the examples or other metadata.

The dataset consists of two rounds, each with a train/dev/test split:

Round 1: Naturally occurring sentences

  • dynasent-v1.1-round01-yelp-train.jsonl
  • dynasent-v1.1-round01-yelp-dev.jsonl
  • dynasent-v1.1-round01-yelp-test.jsonl

Round 1: Sentences crowdsourced using Dynabench

  • dynasent-v1.1-round02-dynabench-train.jsonl
  • dynasent-v1.1-round02-dynabench-dev.jsonl
  • dynasent-v1.1-round02-dynabench-test.jsonl

SST-dev revalidation

The dataset also contains a version of the Stanford Sentiment Treebank dev set in our format with labels from our validation task:

  • sst-dev-validated.jsonl

Quick start

This function can be used to load any subset of the files:

import json

def load_dataset(*src_filenames, labels=None):
    data = []
    for filename in src_filenames:
        with open(filename) as f:
            for line in f:
                d = json.loads(line)
                if labels is None or d['gold_label'] in labels:
                    data.append(d)
    return data

For example, to create a Round 1 train set restricting to examples with ternary gold labels:

import os

r1_train_filename = os.path.join('dynasent-v1.1', 'dynasent-v1.1-round01-yelp-train.jsonl')

ternary_labels = ('positive', 'negative', 'neutral')

r1_train = load_dataset(r1_train_filename, labels=ternary_labels)

X_train, y_train = zip(*[(d['sentence'], d['gold_label']) for d in r1_train])

Data format

Round 1 format

{'hit_ids': ['y5238'],
 'sentence': 'Roto-Rooter is always good when you need someone right away.',
 'indices_into_review_text': [0, 60],
 'model_0_label': 'positive',
 'model_0_probs': {'negative': 0.01173639390617609,
  'positive': 0.7473671436309814,
  'neutral': 0.24089649319648743},
 'text_id': 'r1-0000001',
 'review_id': 'IDHkeGo-nxhqX4Exkdr08A', 
 'review_rating': 1,
 'label_distribution': {'positive': ['w130', 'w186', 'w207', 'w264', 'w54'],
  'negative': [],
  'neutral': [],
  'mixed': []},
 'gold_label': 'positive'}

Details:

  • 'hit_ids': List of Amazon Mechanical Turk Human Interface Tasks (HITs) in which this example appeared during validation. The values are anonymized but used consistently throughout the dataset.
  • 'sentence': The example text.
  • 'indices_into_review_text': indices of 'sentence' into the original review in the Yelp Academic Dataset.
  • 'model_0_label': prediction of Model 0 as described in the paper. The possible values are 'positive', 'negative', and 'neutral'.
  • 'model_0_probs': probability distribution predicted by Model 0. The keys are ('positive', 'negative', 'neutral') and the values are floats.
  • 'text_id': unique identifier for this entry.
  • 'review_id': review-level identifier for the review from the Yelp Academic Dataset containing 'sentence'.
  • 'review_rating': review-level star-rating for the review containing 'sentence' in the Yelp Academic Dataset. The possible values are 1, 2, 3, 4, and 5.
  • 'label_distribution': response distribution from the MTurk validation task. The keys are ('positive', 'negative', 'neutral') and the values are lists of anonymized MTurk ids, which are used consistently throughout the dataset.
  • 'gold_label': the label chosen by at least three of the five workers if there is one (possible values: 'positive', 'negative', 'neutral', and 'mixed'), else None.

Here is some code one could use to augment a dataset, as loaded by load_dataset, with a field giving the full review text from the Yelp Academic Dataset:

import json

def index_yelp_reviews(yelp_src_filename='yelp_academic_dataset_review.json'):
    index = {}
    with open(yelp_src_filename) as f:
        for line in f:
            d = json.loads(line)
            index[d['review_id']] = d['text']
    return index

yelp_index = index_yelp_reviews()

def add_review_text_round1(dataset, yelp_index):
    for d in dataset:
        review_text = yelp_index[d['text_id']]
        # Check that we can find the sentence as expected:
        start, end = d['indices_into_review_text']
        assert review_text[start: end] == d['sentence']
        d['review_text'] = review_text
    return dataset

Round 2 format

{'hit_ids': ['y22661'],
 'sentence': "We enjoyed our first and last meal in Toronto at Bombay Palace, and I can't think of a better way to book our journey.",
 'sentence_author': 'w250',
 'has_prompt': True,
 'prompt_data': {'indices_into_review_text': [2093, 2213],
  'review_rating': 5,
  'prompt_sentence': "Our first and last meals in Toronto were enjoyed at Bombay Palace and I can't think of a better way to bookend our trip.",
  'review_id': 'Krm4kSIb06BDHternF4_pA'},
 'model_1_label': 'positive',
 'model_1_probs': {'negative': 0.29140257835388184,
  'positive': 0.6788994669914246,
  'neutral': 0.029697999358177185},
 'text_id': 'r2-0000001',
 'label_distribution': {'positive': ['w43', 'w26', 'w155', 'w23'],
  'negative': [],
  'neutral': [],
  'mixed': ['w174']},
 'gold_label': 'positive'}

Details:

  • 'hit_ids': List of Amazon Mechanical Turk Human Interface Tasks (HITs) in which this example appeared during validation. The values are anonymized but used consistently throughout the dataset.
  • 'sentence': The example text.
  • 'sentence_author': Anonymized MTurk id of the worker who wrote 'sentence'. These are from the same family of ids as used in 'label_distribution', but this id is never one of the ids in 'label_distribution' for this example.
  • 'has_prompt': True if the 'sentence' was written with a Prompt else False.
  • 'prompt_data': None if 'has_prompt' is False, else:
    • 'indices_into_review_text': indices of 'prompt_sentence' into the original review in the Yelp Academic Dataset.
    • 'review_rating': review-level star-rating for the review containing 'sentence' in the Yelp Academic Dataset.
    • 'prompt_sentence': The prompt text.
    • 'review_id': review-level identifier for the review from the Yelp Academic Dataset containing 'prompt_sentence'.
  • 'model_1_label': prediction of Model 1 as described in the paper. The possible values are 'positive', 'negative', and 'neutral'.
  • 'model_1_probs': probability distribution predicted by Model 1. The keys are ('positive', 'negative', 'neutral') and the values are floats.
  • 'text_id': unique identifier for this entry.
  • 'label_distribution': response distribution from the MTurk validation task. The keys are ('positive', 'negative', 'neutral') and the values are lists of anonymized MTurk ids, which are used consistently throughout the dataset.
  • 'gold_label': the label chosen by at least three of the five workers if there is one (possible values: 'positive', 'negative', 'neutral', and 'mixed'), else None.

To add the review texts to the 'prompt_data' field, one can extend the code above for Round 1 with the following function:

def add_review_text_round2(dataset, yelp_index):
    for d in dataset:
        if d['has_prompt']:
            prompt_data = d['prompt_data']
            review_text = yelp_index[prompt_data['review_id']]
            # Check that we can find the sentence as expected:
            start, end = prompt_data['indices_into_review_text']
            assert review_text[start: end] == prompt_data['prompt_sentence']
            prompt_data['review_text'] = review_text
    return dataset

SST-dev format

{'hit_ids': ['s20533'],
 'sentence': '-LRB- A -RRB- n utterly charming and hilarious film that reminded me of the best of the Disney comedies from the 60s.',
 'tree': '(4 (2 (1 -LRB-) (2 (2 A) (3 -RRB-))) (4 (4 (2 n) (4 (3 (2 utterly) (4 (3 (4 charming) (2 and)) (4 hilarious))) (3 (2 film) (3 (2 that) (4 (4 (2 (2 reminded) (3 me)) (4 (2 of) (4 (4 (2 the) (4 best)) (2 (2 of) (3 (2 the) (3 (3 Disney) (2 comedies))))))) (2 (2 from) (2 (2 the) (2 60s)))))))) (2 .)))',
 'text_id': 'sst-dev-validate-0000437',
 'sst_label': '4',
 'label_distribution': {'positive': ['w207', 'w3', 'w840', 'w135', 'w26'],
  'negative': [],
  'neutral': [],
  'mixed': []},
 'gold_label': 'positive'}

Details:

  • 'hit_ids': List of Amazon Mechanical Turk Human Interface Tasks (HITs) in which this example appeared during validation. The values are anonymized but used consistently throughout the dataset.
  • 'sentence': The example text.
  • 'tree': The parsetree for the example as given in the SST distribution.
  • 'text_id': A new identifier for this example.
  • 'sst_label': The root-node label from the SST. Possible values '0', '1' '2', '3', and '4'.
  • 'label_distribution': response distribution from the MTurk validation task. The keys are ('positive', 'negative', 'neutral') and the values are lists of anonymized MTurk ids, which are used consistently throughout the dataset.
  • 'gold_label': the label chosen by at least three of the five workers if there is one (possible values: 'positive', 'negative', 'neutral', and 'mixed'), else None.

Models

Model 0 and Model 1 from the paper are available here:

https://drive.google.com/drive/folders/1dpKrjNJfAILUQcJPAFc5YOXUT51VEjKQ?usp=sharing

This repository includes a Python module dynasent_models.py that provides a Hugging Face-based wrapper around these (PyTorch) models. Simple examples:

import os
from dynasent_models import DynaSentModel

# `dynasent_model0` should be downloaded from the above Google Drive link and 
# placed in the `models` directory. `dynasent_model1` works the same way.
model = DynaSentModel(os.path.join('models', 'dynasent_model0.bin'))

examples = [
    "superb",
    "They said the experience would be amazing, and they were right!",
    "They said the experience would be amazing, and they were wrong!"]

model.predict(examples)

This should return the list ['positive', 'positive', 'negative'].

The predict_proba method provides access to the predicted distribution over the class labels; see the demo at the bottom of dynasent_models.py for details.

The following code uses load_dataset from above to reproduce the Round 2 dev-set report on Model 0 from the paper:

import os
from sklearn.metrics import classification_report
from dynasent_models import DynaSentModel

dev_filename = os.path.join('dynasent-v1.1', 'dynasent-v1.1-round02-dynabench-dev.jsonl')

dev = load_dataset(dev_filename)

X_dev, y_dev = zip(*[(d['sentence'], d['gold_label']) for d in dev])

model = DynaSentModel(os.path.join('models', 'dynasent_model0.bin'))

preds = model.predict(X_dev)

print(classification_report(y_dev, preds, digits=3))

For a fuller report on these models, see our paper and our model card.

Other files

Analysis notebooks

The following notebooks reproduce the dataset statistics, figures, and random example selections from the paper:

  • analyses_comparative.ipynb
  • analysis_round1.ipynb
  • analysis_round2.ipynb
  • analysis_sst_dev_revalidate.ipynb

The Python module dynasent_utils.py contains functions that support those notebooks, and dynasent.mplstyle helps with styling the plots.

Datasheet

The Datasheet for our dataset:

Model Card

The Model Card for our models:

Tests

The module test_dataset.py contains PyTest tests for the dataset. To use it, run

py.test -vv test_dataset.py

in the root directory of this repository.

Validation HIT code

The file validation-hit-contents.html contains the HTML/Javascript used in the validation task. It could be used directly on Amazon Mechanical Turk, by simply pasting its contents into the usual HIT creation window.

License

DynaSent has a Creative Commons Attribution 4.0 International License.

Edit dataset card
Evaluate models HF Leaderboard