guyhadad01 commited on
Commit
ee49a09
·
verified ·
1 Parent(s): 6456c17

Training in progress, step 200, checkpoint

Browse files
last-checkpoint/1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
last-checkpoint/README.md ADDED
@@ -0,0 +1,706 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - dense
7
+ - generated_from_trainer
8
+ - dataset_size:122875
9
+ - loss:CachedMultipleNegativesRankingLoss
10
+ base_model: sentence-transformers/all-MiniLM-L6-v2
11
+ widget:
12
+ - source_sentence: "title: \nBaby Formula Dispenser Portable Milk Powder Dispenser\
13
+ \ Snack Storage Container for Travel Bedroom Outdoor Large Pink"
14
+ sentences:
15
+ - 'description
16
+
17
+ Unique Design Protable handle Child Lock Hollow Scraper Scoop Features Compact
18
+ Size Lightweight Package included 1 x Formula DispenserSize of 4x4x42 this formula
19
+ container can hold about 30 scoops formula for your baby BPAFree
20
+
21
+ Compact Size the formula carrier is small and compact but holds a lot It fits
22
+ in your diaper bag organizer perfectly
23
+
24
+ Easy for OntheGo this baby formula holder makes feeding time easier when you and
25
+ baby are out and about
26
+
27
+ Unique ChildLock and silicone sealing ring ensure airtightness and does not leak
28
+ in any time
29
+
30
+ Multipurpose it also can be used as snack cups or container for storing oatmeal
31
+ cornflour nuts candy fruit'
32
+ - 'description
33
+
34
+ Stay on top of diaper odor and cleanup with the upup 1Pack Diaper Pail Refills
35
+ Theyre compatible with Genie Elite Diaper Genie II Elite Diaper Genie Essentials
36
+ and Diaper Genie II pails The refills are made with a 7layer film that keeps odor
37
+ at bay so the nursery stays fresh Holds up to 272 newborn diapers Quality needs
38
+ priced to please With up up your satisfaction is 100 guaranteed or your money
39
+ back Features 7layer Plastic Material Plastic Care and Cleaning Does Not Apply
40
+ Dimensions 1950H x 6550W x 6550L Imported'
41
+ - 'description
42
+
43
+ This super soft wearable blanket replaces loose blankets in the crib for a safer
44
+ sleeping environment65 Cotton 35 Polyester
45
+
46
+ Imported
47
+
48
+ Machine Wash
49
+
50
+ Includes 1 zipup wearable blanket neck widths 36 mos 3 618 mos 35 1824 mos 4
51
+
52
+ Certified free from over 100 harmful chemicals by OekoTex Standard 100
53
+
54
+ An Amazon Brand
55
+
56
+ Fill is made from 100 polyester
57
+
58
+ Handmedown lasting quality
59
+
60
+ Prewashed fabric for added softness and minimal shrinkage'
61
+ - source_sentence: "title: \nDream On Me 2in1 Folding Birch Portable Crib Cherry"
62
+ sentences:
63
+ - 'description
64
+
65
+ Clean and pamper your baby with the Lil Luxuries Whirlpool Bubbling Spa and Shower
66
+ in a luxurious spa experience at home With a calming whirlpool and massaging bubbles
67
+ keeping your little one entertained and soothed during bath time has never been
68
+ easier The innovative design of this baby bathtub keeps baby warm during bath
69
+ time with the insulated double wall tub For added convenience and comfort the
70
+ deluxe newborn sling with plush bolster can be used in the sink infant tub or
71
+ adult bathtub Keep baby entertained with the motorized water jet by creating bubbles
72
+ and vibrations for the ultimate spa effect This baby tubs convenient design grows
73
+ with your little one as the removable clean rinse spashower unit can be used in
74
+ adult tub for extended useLUXURIOUS BABY BATHTUB Clean and pamper your baby with
75
+ the Lil Luxuries Whirlpool Bubbling Spa and Shower in a luxurious spa experience
76
+ at home Keep baby soothed with the calming circulating water jets
77
+
78
+ KEEPS NEWBORNS SUPPORTED The deluxe newborn sling with plush bolster can be used
79
+ in the sink infant tub or adult bathtub
80
+
81
+ KEEPS BABY WARM The innovative design of this baby bathtub keeps baby warm during
82
+ bath time with the insulated double wall tub
83
+
84
+ KEEP BABY ENTERTAINED The motorized water jet keeps baby entertained by creating
85
+ bubbles and vibrations for the ultimate spa effect
86
+
87
+ GROWS WITH BABY This baby tubs convenient design grows with your little one The
88
+ removable clean rinse spashower unit can be used in adult tub for extended use'
89
+ - 'description
90
+
91
+ The Dream On Me 2in1 Folding Birch Portable Crib is a must for todays urban onthego
92
+ parents and designed to make the most of small spaces The crib assembles in a
93
+ jiffy and the US patented rail design easily helps convert the crib into a playpen
94
+ in seconds It makes for the perfect spacesaver with the unique hinge system allowing
95
+ it to fold flat for compact storage Its lightweight portable frame with wheels
96
+ makes it easy for you to move it around the house Take it on your travels or to
97
+ grandmas house Safety is a top priority at Dream On Me The stationary non drop
98
+ side rail system and heavy duty locking commercial size wheels provide the utmost
99
+ in product safety It meets the ASTM standards It features a solid birchwood frame
100
+ Its clean minimalist design and four beautiful finish options allow it to blend
101
+ into any room decor Crib dimensions are 40L x 26W x 38H Comes with a twoposition
102
+ mattress support One inch crib mattress and tools for easy assembly includedUSEFUL
103
+ CONVERTIBILITY Perfect for todays urban parents with its dualfunctionality Crib
104
+ assembly is quick and the US patented rail design easily helps convert crib into
105
+ a playpen in seconds
106
+
107
+ COMPACT PORTABLE The perfect spacesaver unique hinge system allows it to fold
108
+ flat for compact storage making it ideal for small space Lightweight portable
109
+ frame with wheels make it the ideal portable travel companion
110
+
111
+ SAFETY MATTERS The stationary non drop side rail system and heavy duty locking
112
+ commercial size wheels provide the utmost in product safety It meets the ASTM
113
+ standards
114
+
115
+ CLEAN DESIGN Comes with a solid birchwood frame Clean minimalist design and four
116
+ beautiful finish options allow it to blend into any room decor One inch crib mattress
117
+ and tools for easy assembly included
118
+
119
+ PRODUCT SPECS Crib dimensions are 40L x 26W x 38H Comes with a twoposition mattress
120
+ support system'
121
+ - 'description
122
+
123
+ Johnsons Cotton Swabs
124
+
125
+ are made from only 100 pure nonchlorine bleached cotton  Pure cotton is naturally
126
+ soft absorbent and gentle  Give your skin the gentle natural care of Johnsons
127
+ Cotton productOur 100 pure high quality cotton has just the right combination
128
+ of softness and strength
129
+
130
+ Here are some ways you can use JOHNSONS Pure Cotton Swabs Beauty Care Baby Care
131
+ Crafts Household Tasks'
132
+ - source_sentence: "title: \nEUBEST Baby Mosquito Net Baby Toddler Bed Crib Canopy\
133
+ \ Netting White"
134
+ sentences:
135
+ - 'description
136
+
137
+ Product Description
138
+
139
+ A customized fit is always better than a standard one especially when it comes
140
+ to babyproofing If you need to extend your Munchkin safety gate you can do so
141
+ with standard gate extensions available in a variety of sizes and colors to give
142
+ you the perfect fit These extensions work with pressurefit safety gates and can
143
+ be combined to span a variety of opening sizes up to 54 wide Available in 275
144
+ 55 and 11 sizes For use with the AutoClose Metal Gate MK0006 Easy Close Metal
145
+ Gate MK0002 and Safe Step with TripGuard Gate MK0096
146
+
147
+ From the Manufacturer
148
+
149
+ Add extra width to your Munchkin AutoClose or Easy Close gate Simply measure the
150
+ narrowest point of your opening often at the base board and select the dimensions
151
+ that correspond to your measurements
152
+
153
+ Manufacturer Contact Information
154
+
155
+ Phone 1800344BABY 2229 Email cservmunchkincom
156
+
157
+ See moreExtends the AutoClose Metal Gate MK0006 Easy Close Metal Gate MK0002 and
158
+ Safe Step with TripGuard Gate MK0096
159
+
160
+ Dimensions 11 W x 295 H
161
+
162
+ Steel extensions can be combined to span a variety of openings up to 54
163
+
164
+ Material Steel'
165
+ - 'description
166
+
167
+ Your baby deserves the best
168
+
169
+ We believe that there is nothing more precious delicate and beautiful than your
170
+ baby Qz creation with children AND families in mind Whether you are looking for
171
+ a baby gift a birthday present or family gift hope you find the perfect baby Muslin
172
+ Burp Cloths that will grow with your family
173
+
174
+ Carefully made from GOTS verified organic cotton
175
+
176
+ The Global Organic Textile Standard GOTS is recognised as the worlds leading processing
177
+ standard for textiles made from organic fibres Organic cotton is grown without
178
+ the use of chemical fertilizers herbicides or pesticides It is hypoallergenic
179
+ and soft next to babys sensitive skin
180
+
181
+ Muslin is a finely woven cotton open weave cotton that is lightweight breathable
182
+ and softens with every wash
183
+
184
+ Qz Baby Guarantee
185
+
186
+ Offering the very best for your baby is the most important thing in the world
187
+ right We are confident that we can do exactly that for you If you are unsatisfied
188
+ for any reason contact us for a refund Get this ultrasoft organic cotton muslin
189
+ burp cloth today You will not regret investing in a highquality burp cloth for
190
+ your baby
191
+
192
+ Click the Add to Cart button NOW and get your very own Qz Organic Cotton Muslin
193
+ Burp Cloth TODAY FOR 32 OFFOrganic Cotton
194
+
195
+ SAFER 100 Organic materials with GOTS certified Muslin in natural white and ultimately
196
+ packed in the premium muslin bag our baby burp bibs are free from harmful chemicals
197
+ dyes which makes them safer to baby and our planet Labels are sewn appropriately
198
+ so that babys skin doesnt get hurts Made with fully certified organic cotton that
199
+ is grown and manufactured in an environmentally responsible way NO AZO NICKEL
200
+ FORMALDEHYDE on all our muslin burp cloth
201
+
202
+ BEST BURP CLOTHS BABY BIBS BABY CAPE A bib burp cloth and cape in one our 4 pack
203
+ of soft burp cloths keeps you and your little one clean Drape it over your babys
204
+ shoulders and snap it at the back for a fullcoverage bib or using as a cape when
205
+ in aircondition and windy day or place it over your shoulder for a noslip burp
206
+ cloth Perfect for teething or drooling babies
207
+
208
+ EIGHT ABSORBENT THICK and THIRSTY LAYERS are generously sized noslip to absorb
209
+ dribbles lightweight and yet breathable to keep baby cozy dry and comfy Babys
210
+ skin will feel soft and smooth next to organically grown cotton Muslin Burp Cloths
211
+
212
+ REUSABLE Most Absorbent and Most Durable burp cloth Better than cotton or terry
213
+ cloth they actually get SOFTER after each wash Reusable so you can use them Over Over
214
+ Again It can save your money More economical and environmental Lets Use as a burp
215
+ cloth now and as a wonderfully soft washcloth later
216
+
217
+ OVERSIZED Our prewashed burp cloths come in a generous 25x58cm size 2cm providing
218
+ greater shoulder coverage and less outfit changes for Mom and Dad'
219
+ - 'description
220
+
221
+ Pacakge1pcsThis mosquito net fits playpens bassinets cribs smaller beds etc It
222
+ would add elegance to the room and at the same time protect your baby from mosquitoes
223
+ and bugs
224
+
225
+ Perfect choice for your babys room So cute
226
+
227
+ Easier match for the bed L W 150 80 cm 59 315 inch
228
+
229
+ Lightweight in good condition and in great shape
230
+
231
+ shipping from Uhome'
232
+ - source_sentence: "title: \nLittle Seeds Rowan Valley Flint 6Drawer Dresser BlackWhite"
233
+ sentences:
234
+ - 'description
235
+
236
+ BisphenolA BPA is used in hard clear plastics such as baby bottles to prevent
237
+ breakage In 2006 when we learned that BPA also disrupts reproductive functions
238
+ and causes other developmental problems we discontinued all BPA from our feeding
239
+ items We know that lugging around everything parents need for their little ones
240
+ can be a real chore We have a collection of spillproof easy to carry feeding items
241
+ to make life simpler when onthegoCute lunchbox with locking lid
242
+
243
+ Nesting divided tray and bowl
244
+
245
+ Dimensions 6x 475x 275152 x 12 x 7 cm
246
+
247
+ BPA Free'
248
+ - 'description
249
+
250
+ Add a modern look to your childs bedroom with the Little Seeds Rowan Valley Flint
251
+ 6 Drawer Dresser Made from sturdy MDF the nontoxic painted matte black finish
252
+ on the chassis contrasts the white finish of the drawer fronts for a clean sleek
253
+ look The solid wood angled feet give the Dresser a stable base and adds a decorative
254
+ touch The metal slides on each drawer have builtin stops for safety and can withstand
255
+ years of use Keep your childs sweaters folded shirts and pants and seasonal coats
256
+ and gloves stored away and organized in the 6 drawers This 6 Drawer Dresser can
257
+ convert to a changing table by adding our optional changing station top sold separately
258
+ The 6 Drawer Dresser meets or exceeds the CPSIA Juvenile testing requirements
259
+ and comes with a wall anchor kit to ensure your childs safety The Little Seeds
260
+ Rowan Valley Flint 6 Drawer Dresser ships flat to your door and requires assembly
261
+ upon opening Two adults are recommended for assembly Once assembled the Dresser
262
+ measures to be 32625H x 47625W x 17875D Little Seeds not only creates this and
263
+ many more on trend baby to teen furniture pieces we also partner with various
264
+ environmental protection programs to protect pollinator and other wildlife habitats
265
+ for future generationsThe Little Seeds Rowan Valley Flint 6 Drawer Dresser gives
266
+ your childs bedroom a modern updated look
267
+
268
+ Made from sturdy MDF the painted nontoxic finish gives the Dresser an updated
269
+ clean look The solid wood angled feet offer a decorative touch that adds stability
270
+ Each drawer has metal slides with builtin stops for safety
271
+
272
+ The Dresser can convert to a Changing Table by adding the changing station topper
273
+ sold separately The 6 Drawer Dresser meets or exceeds the CPSIA Juvenile testing
274
+ requirements and comes with a wall anchor kit to ensure your childs safety
275
+
276
+ Complete your childs bedroom with the Rowan Valley Flint Crib or mix and match
277
+ with other Rowan Valley collections for an eclectic look
278
+
279
+ The Little Seeds Rowan Valley Flint 6 Drawer Dresser ships flat to your door and
280
+ 2 adults are recommended for assembly The Dresser top can hold up to 75 lbs and
281
+ each drawer will hold 35 lbs Assembled dimensions 32625H x 47625W x 17875D'
282
+ - 'description
283
+
284
+ Born Free BPAFree Drinking Cups are designed to help baby smoothly transition
285
+ from bottle to cup  The cups are completely compatible and interchangeable with
286
+ Born Free bottles Mix and match cup handles bottle nipples and soft spouts to
287
+ help baby make and easier transition  The patented ActiveFlow Venting Technology
288
+ is clinically designed to help reduce colic symptoms such as gas and fussiness
289
+ and can help extend breastfeeding by mimicking mothers natural flow which is controlled
290
+ by baby while nursingNatural transition from bottle to cup with a teething friendly
291
+ durable soft spout
292
+
293
+ Natural flow with ActiveFlow Venting Technology
294
+
295
+ Naturally free from BPA and PVC
296
+
297
+ Top rack dishwasher safe
298
+
299
+ All handles collars nipples and spouts are completely interchangeable between
300
+ any Born Free bottle or transition cup'
301
+ - source_sentence: "title: \nDream On Me 4 Piece Toddler Bedding Set Spring Garden"
302
+ sentences:
303
+ - 'description
304
+
305
+ The Baby Jogger City Mini GT2 Stroller Barre Collection has an allnew design with
306
+ forever air rubber tires and allwheel suspension providing uncompromised agility
307
+ on any terrain This special edition fashion includes premium leatherette detailing
308
+ on the canopy and harness covers and an upgraded seat pad for a sleek modern look
309
+ Plus the included belly bar gives your child a safe and comfortable area to hold
310
+ onto while strolling The signature onehand compact fold adjustable handlebar and
311
+ handoperated parking brake provide you the ultimate convenience while the adjustable
312
+ calf support and nearflat seat recline allow you to find the most comfortable
313
+ position for your child Create a travel system using the included Baby Jogger
314
+ infant car seat adapters car seat sold separately or customize your ride with
315
+ a variety of accessoriesNylon
316
+
317
+ Stroller has forever air rubber tires with allwheel suspension providing uncompromised
318
+ agility on any terrain
319
+
320
+ Lift a strap with one hand and the City Mini GT2 stroller folds itself simply
321
+ and compactlyDo not use bleach solvents or abrasive cleaning materials
322
+
323
+ Premium leatherette detailing on the canopy and harness covers and an upgraded
324
+ seat pad for a sleek modern look
325
+
326
+ Included belly bar gives your child a safe and comfortable area to hold onto while
327
+ strolling
328
+
329
+ Add a Baby Jogger Infant Car Seat using the included infant car seat adapters
330
+ car seat sold separately and go anywhere from day one'
331
+ - 'description
332
+
333
+ Let the little one in your home settle down to sleep in this incredible Toddler
334
+ Bed set This girl bedding set features stunning floral print with a butterfly
335
+ combined with bright solid microfiber fabrics This collection uses the stylish
336
+ colors of purple green and purple This set includes comforter fitted Sheet flat
337
+ sheet Pillow caseCottonPolyester
338
+
339
+ 4piece crib bedding set includes a Comforter Flat Sheet Fitted Sheet and Pillowcase
340
+
341
+ Flat Sheet Fitted Sheet and Pillowcase has 35 percent cotton and 65 percent polyester
342
+ fabrics Comforter has 100 percent polyester fiber fill
343
+
344
+ Measures 52 X 28 inches Sheets fit a standard 52 X 28 cribToddler Mattress
345
+
346
+ Machine wash 4piece Toddler Bedding set in cold water gentle cycle
347
+
348
+ Use only nonchlorine bleach when needed tumble dry low heat iron on low heat when
349
+ needed'
350
+ - 'description
351
+
352
+ Hudson Baby plush security blanket set is soft and cuddly and perfect to aid in
353
+ child development Hudson Baby animal face security blanket will keep baby calm
354
+ and content with their favorite animal friend Our security blanket features super
355
+ soft and luxurious fabric with a fun animal creature to play and snuggle to their
356
+ little hearts desire Our security blanket will surely become a baby favorite The
357
+ larger blanket is generously sized perfect for snuggling cuddling swaddling or
358
+ as a car seat or stroller coverSet includes security blanket and plush blanket
359
+
360
+ Made with 100 cozy plush fabric Blanket 30x36 inches and security blanket 14x14
361
+ inches
362
+
363
+ Soft and gentle on babys skin
364
+
365
+ Optimal for everyday use
366
+
367
+ Affordable high quality nursery blanket set'
368
+ datasets:
369
+ - guyhadad01/Amazon_2023_items_processed_filtered
370
+ pipeline_tag: sentence-similarity
371
+ library_name: sentence-transformers
372
+ ---
373
+
374
+ # SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
375
+
376
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) on the [amazon_2023_items_processed_filtered](https://huggingface.co/datasets/guyhadad01/Amazon_2023_items_processed_filtered) dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
377
+
378
+ ## Model Details
379
+
380
+ ### Model Description
381
+ - **Model Type:** Sentence Transformer
382
+ - **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
383
+ - **Maximum Sequence Length:** 512 tokens
384
+ - **Output Dimensionality:** 384 dimensions
385
+ - **Similarity Function:** Cosine Similarity
386
+ - **Training Dataset:**
387
+ - [amazon_2023_items_processed_filtered](https://huggingface.co/datasets/guyhadad01/Amazon_2023_items_processed_filtered)
388
+ <!-- - **Language:** Unknown -->
389
+ <!-- - **License:** Unknown -->
390
+
391
+ ### Model Sources
392
+
393
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
394
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers)
395
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
396
+
397
+ ### Full Model Architecture
398
+
399
+ ```
400
+ SentenceTransformer(
401
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
402
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
403
+ (2): Normalize()
404
+ )
405
+ ```
406
+
407
+ ## Usage
408
+
409
+ ### Direct Usage (Sentence Transformers)
410
+
411
+ First install the Sentence Transformers library:
412
+
413
+ ```bash
414
+ pip install -U sentence-transformers
415
+ ```
416
+
417
+ Then you can load this model and run inference.
418
+ ```python
419
+ from sentence_transformers import SentenceTransformer
420
+
421
+ # Download from the 🤗 Hub
422
+ model = SentenceTransformer("guyhadad01/EncodeRec_22M_Baby")
423
+ # Run inference
424
+ sentences = [
425
+ 'title: \nDream On Me 4 Piece Toddler Bedding Set Spring Garden',
426
+ 'description\nLet the little one in your home settle down to sleep in this incredible Toddler Bed set This girl bedding set features stunning floral print with a butterfly combined with bright solid microfiber fabrics This collection uses the stylish colors of purple green and purple This set includes comforter fitted Sheet flat sheet Pillow caseCottonPolyester\n4piece crib bedding set includes a Comforter Flat Sheet Fitted Sheet and Pillowcase\nFlat Sheet Fitted Sheet and Pillowcase has 35 percent cotton and 65 percent polyester fabrics Comforter has 100 percent polyester fiber fill\nMeasures 52 X 28 inches Sheets fit a standard 52 X 28 cribToddler Mattress\nMachine wash 4piece Toddler Bedding set in cold water gentle cycle\nUse only nonchlorine bleach when needed tumble dry low heat iron on low heat when needed',
427
+ 'description\nHudson Baby plush security blanket set is soft and cuddly and perfect to aid in child development Hudson Baby animal face security blanket will keep baby calm and content with their favorite animal friend Our security blanket features super soft and luxurious fabric with a fun animal creature to play and snuggle to their little hearts desire Our security blanket will surely become a baby favorite The larger blanket is generously sized perfect for snuggling cuddling swaddling or as a car seat or stroller coverSet includes security blanket and plush blanket\nMade with 100 cozy plush fabric Blanket 30x36 inches and security blanket 14x14 inches\nSoft and gentle on babys skin\nOptimal for everyday use\nAffordable high quality nursery blanket set',
428
+ ]
429
+ embeddings = model.encode(sentences)
430
+ print(embeddings.shape)
431
+ # [3, 384]
432
+
433
+ # Get the similarity scores for the embeddings
434
+ similarities = model.similarity(embeddings, embeddings)
435
+ print(similarities)
436
+ # tensor([[ 1.0000, 0.5316, -0.0116],
437
+ # [ 0.5316, 1.0000, 0.0884],
438
+ # [-0.0116, 0.0884, 1.0000]])
439
+ ```
440
+
441
+ <!--
442
+ ### Direct Usage (Transformers)
443
+
444
+ <details><summary>Click to see the direct usage in Transformers</summary>
445
+
446
+ </details>
447
+ -->
448
+
449
+ <!--
450
+ ### Downstream Usage (Sentence Transformers)
451
+
452
+ You can finetune this model on your own dataset.
453
+
454
+ <details><summary>Click to expand</summary>
455
+
456
+ </details>
457
+ -->
458
+
459
+ <!--
460
+ ### Out-of-Scope Use
461
+
462
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
463
+ -->
464
+
465
+ <!--
466
+ ## Bias, Risks and Limitations
467
+
468
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
469
+ -->
470
+
471
+ <!--
472
+ ### Recommendations
473
+
474
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
475
+ -->
476
+
477
+ ## Training Details
478
+
479
+ ### Training Dataset
480
+
481
+ #### amazon_2023_items_processed_filtered
482
+
483
+ * Dataset: [amazon_2023_items_processed_filtered](https://huggingface.co/datasets/guyhadad01/Amazon_2023_items_processed_filtered) at [6b58dd1](https://huggingface.co/datasets/guyhadad01/Amazon_2023_items_processed_filtered/tree/6b58dd18854109aac31652e941c667725f6352f0)
484
+ * Size: 122,875 training samples
485
+ * Columns: <code>title</code> and <code>description</code>
486
+ * Approximate statistics based on the first 1000 samples:
487
+ | | title | description |
488
+ |:--------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
489
+ | type | string | string |
490
+ | details | <ul><li>min: 6 tokens</li><li>mean: 23.22 tokens</li><li>max: 51 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 237.75 tokens</li><li>max: 512 tokens</li></ul> |
491
+ * Samples:
492
+ | title | description |
493
+ |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
494
+ | <code>title: <br>Chicco Viaro Travel System Teak</code> | <code>description<br>Product Description<br>For ultimate convenience the Chicco Viaro QuickFold Stroller has a sleek threewheel design lightweight aluminum frame and onehand quick fold A pullstrap and button are conveniently tucked under the seat and easy to activate simultaneously for a compact freestanding fold The stroller is even easier to open again after closingFor infants the Viaro Stroller functions as a travel system with easy clickin attachment for the KeyFit 30 Infant Car Seat For older riders the Viaro Stroller includes a detachable tray with two cup holders adjustable canopy and multiposition backrest A swiveling front wheel and suspension help maintain a smooth ride from surface to surface Toetap rear brakes keep the stroller in place when parked For parents the Viaro Stroller features a padded pushhandle parent tray with two cup holders and a large basket that is easily accessible from the front or back The Viaro Travel System includes the 1rated Chicco KeyFit 30 Infant Car Seat wh...</code> |
495
+ | <code>title: <br>Kisbaby Four Layers Muslin Lightweight Unisex Toddler Wearable Blanket Early Walker Sleeping Sack Footed Sleep Sack for Baby Girls and Boys Blue Star 12M</code> | <code>description<br>You can choose bigger size If you confuse about size<br>Medium Age 612 months Height up to 32inches Weight up to 24lbs<br>Large Age 1218 months Height up to 35inches Weight up to 28lbs<br>Kids Cant Kick Off and Stay Covered All Night<br>Super soft draped muslin fabric<br>Lightweight and breathable keep baby warm and reduces the risk of overheating<br>Tuck feet in at night for warmth<br>A blanket your little one cant kick off ensuring better more comfortable sleep<br>Foot opening for easy mobility<br>Foot opening provide greater mobility when your little one is awake<br>Easy bathroom visits<br>Inverted zipper for easy bathroom visits or diaper changing95 Cotton 4 Layer Muslin Hand Wash in Cold Water is Recommended<br>Imported<br>Draped muslin softer touch lightweight breathable and friendly to baby TOG 10<br>Use over regular sleepwear to replace loose blankets cannot be kicked off so baby stays warm all night<br>Short sleeve for ease of dressing and freedom of movement reduces the risk of overheating<br>Unzips from the bo...</code> |
496
+ | <code>title: <br>Nuby iMonster Toddler Bowl</code> | <code>description<br>When babies begin to show interest in feeding themselves having the right equipment is a must Nuby feeding accessories are made to withstand the challenges of daily use and retain their bright color This Toddler Bowl is ideal for children who are beginning to feed themselves Suitable for warm or cold foods Microwave safe BPA FreeMakes feeding fun for baby and easier for parent<br>Easy Grip design makes it easy for baby to hold<br>Bright color captures your babys attention Fun character inspires your childs imagination<br>Wash prior to initial use and after subsequent uses Easy clean with warm soapy water Rinse and dry thoroughly Also safe for top rack of dishwasher<br>18 Months BPA Free Microwave safe</code> |
497
+ * Loss: [<code>CachedMultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedmultiplenegativesrankingloss) with these parameters:
498
+ ```json
499
+ {
500
+ "scale": 20.0,
501
+ "similarity_fct": "cos_sim",
502
+ "mini_batch_size": 64,
503
+ "gather_across_devices": false
504
+ }
505
+ ```
506
+
507
+ ### Training Hyperparameters
508
+ #### Non-Default Hyperparameters
509
+
510
+ - `per_device_train_batch_size`: 512
511
+ - `num_train_epochs`: 1
512
+ - `warmup_ratio`: 0.1
513
+ - `bf16`: True
514
+ - `push_to_hub`: True
515
+ - `hub_model_id`: guyhadad01/EncodeRec_22M_Baby
516
+ - `hub_strategy`: checkpoint
517
+
518
+ #### All Hyperparameters
519
+ <details><summary>Click to expand</summary>
520
+
521
+ - `overwrite_output_dir`: False
522
+ - `do_predict`: False
523
+ - `eval_strategy`: no
524
+ - `prediction_loss_only`: True
525
+ - `per_device_train_batch_size`: 512
526
+ - `per_device_eval_batch_size`: 8
527
+ - `per_gpu_train_batch_size`: None
528
+ - `per_gpu_eval_batch_size`: None
529
+ - `gradient_accumulation_steps`: 1
530
+ - `eval_accumulation_steps`: None
531
+ - `torch_empty_cache_steps`: None
532
+ - `learning_rate`: 5e-05
533
+ - `weight_decay`: 0.0
534
+ - `adam_beta1`: 0.9
535
+ - `adam_beta2`: 0.999
536
+ - `adam_epsilon`: 1e-08
537
+ - `max_grad_norm`: 1.0
538
+ - `num_train_epochs`: 1
539
+ - `max_steps`: -1
540
+ - `lr_scheduler_type`: linear
541
+ - `lr_scheduler_kwargs`: {}
542
+ - `warmup_ratio`: 0.1
543
+ - `warmup_steps`: 0
544
+ - `log_level`: passive
545
+ - `log_level_replica`: warning
546
+ - `log_on_each_node`: True
547
+ - `logging_nan_inf_filter`: True
548
+ - `save_safetensors`: True
549
+ - `save_on_each_node`: False
550
+ - `save_only_model`: False
551
+ - `restore_callback_states_from_checkpoint`: False
552
+ - `no_cuda`: False
553
+ - `use_cpu`: False
554
+ - `use_mps_device`: False
555
+ - `seed`: 42
556
+ - `data_seed`: None
557
+ - `jit_mode_eval`: False
558
+ - `bf16`: True
559
+ - `fp16`: False
560
+ - `fp16_opt_level`: O1
561
+ - `half_precision_backend`: auto
562
+ - `bf16_full_eval`: False
563
+ - `fp16_full_eval`: False
564
+ - `tf32`: None
565
+ - `local_rank`: 0
566
+ - `ddp_backend`: None
567
+ - `tpu_num_cores`: None
568
+ - `tpu_metrics_debug`: False
569
+ - `debug`: []
570
+ - `dataloader_drop_last`: False
571
+ - `dataloader_num_workers`: 0
572
+ - `dataloader_prefetch_factor`: None
573
+ - `past_index`: -1
574
+ - `disable_tqdm`: False
575
+ - `remove_unused_columns`: True
576
+ - `label_names`: None
577
+ - `load_best_model_at_end`: False
578
+ - `ignore_data_skip`: False
579
+ - `fsdp`: []
580
+ - `fsdp_min_num_params`: 0
581
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
582
+ - `fsdp_transformer_layer_cls_to_wrap`: None
583
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
584
+ - `parallelism_config`: None
585
+ - `deepspeed`: None
586
+ - `label_smoothing_factor`: 0.0
587
+ - `optim`: adamw_torch
588
+ - `optim_args`: None
589
+ - `adafactor`: False
590
+ - `group_by_length`: False
591
+ - `length_column_name`: length
592
+ - `project`: huggingface
593
+ - `trackio_space_id`: trackio
594
+ - `ddp_find_unused_parameters`: None
595
+ - `ddp_bucket_cap_mb`: None
596
+ - `ddp_broadcast_buffers`: False
597
+ - `dataloader_pin_memory`: True
598
+ - `dataloader_persistent_workers`: False
599
+ - `skip_memory_metrics`: True
600
+ - `use_legacy_prediction_loop`: False
601
+ - `push_to_hub`: True
602
+ - `resume_from_checkpoint`: None
603
+ - `hub_model_id`: guyhadad01/EncodeRec_22M_Baby
604
+ - `hub_strategy`: checkpoint
605
+ - `hub_private_repo`: None
606
+ - `hub_always_push`: False
607
+ - `hub_revision`: None
608
+ - `gradient_checkpointing`: False
609
+ - `gradient_checkpointing_kwargs`: None
610
+ - `include_inputs_for_metrics`: False
611
+ - `include_for_metrics`: []
612
+ - `eval_do_concat_batches`: True
613
+ - `fp16_backend`: auto
614
+ - `push_to_hub_model_id`: None
615
+ - `push_to_hub_organization`: None
616
+ - `mp_parameters`:
617
+ - `auto_find_batch_size`: False
618
+ - `full_determinism`: False
619
+ - `torchdynamo`: None
620
+ - `ray_scope`: last
621
+ - `ddp_timeout`: 1800
622
+ - `torch_compile`: False
623
+ - `torch_compile_backend`: None
624
+ - `torch_compile_mode`: None
625
+ - `include_tokens_per_second`: False
626
+ - `include_num_input_tokens_seen`: no
627
+ - `neftune_noise_alpha`: None
628
+ - `optim_target_modules`: None
629
+ - `batch_eval_metrics`: False
630
+ - `eval_on_start`: False
631
+ - `use_liger_kernel`: False
632
+ - `liger_kernel_config`: None
633
+ - `eval_use_gather_object`: False
634
+ - `average_tokens_across_devices`: True
635
+ - `prompts`: None
636
+ - `batch_sampler`: batch_sampler
637
+ - `multi_dataset_batch_sampler`: proportional
638
+ - `router_mapping`: {}
639
+ - `learning_rate_mapping`: {}
640
+
641
+ </details>
642
+
643
+ ### Training Logs
644
+ | Epoch | Step | Training Loss |
645
+ |:------:|:----:|:-------------:|
646
+ | 0.2083 | 50 | 1.1163 |
647
+ | 0.4167 | 100 | 0.6972 |
648
+ | 0.625 | 150 | 0.6743 |
649
+ | 0.8333 | 200 | 0.6641 |
650
+
651
+
652
+ ### Framework Versions
653
+ - Python: 3.12.11
654
+ - Sentence Transformers: 5.1.2
655
+ - Transformers: 4.57.0
656
+ - PyTorch: 2.7.1+cu126
657
+ - Accelerate: 1.10.0
658
+ - Datasets: 3.6.0
659
+ - Tokenizers: 0.22.1
660
+
661
+ ## Citation
662
+
663
+ ### BibTeX
664
+
665
+ #### Sentence Transformers
666
+ ```bibtex
667
+ @inproceedings{reimers-2019-sentence-bert,
668
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
669
+ author = "Reimers, Nils and Gurevych, Iryna",
670
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
671
+ month = "11",
672
+ year = "2019",
673
+ publisher = "Association for Computational Linguistics",
674
+ url = "https://arxiv.org/abs/1908.10084",
675
+ }
676
+ ```
677
+
678
+ #### CachedMultipleNegativesRankingLoss
679
+ ```bibtex
680
+ @misc{gao2021scaling,
681
+ title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
682
+ author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
683
+ year={2021},
684
+ eprint={2101.06983},
685
+ archivePrefix={arXiv},
686
+ primaryClass={cs.LG}
687
+ }
688
+ ```
689
+
690
+ <!--
691
+ ## Glossary
692
+
693
+ *Clearly define terms in order to be accessible across audiences.*
694
+ -->
695
+
696
+ <!--
697
+ ## Model Card Authors
698
+
699
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
700
+ -->
701
+
702
+ <!--
703
+ ## Model Card Contact
704
+
705
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
706
+ -->
last-checkpoint/config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertModel"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "dtype": "bfloat16",
8
+ "gradient_checkpointing": false,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 384,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 1536,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 6,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "transformers_version": "4.57.0",
22
+ "type_vocab_size": 2,
23
+ "use_cache": true,
24
+ "vocab_size": 30522
25
+ }
last-checkpoint/config_sentence_transformers.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "5.1.2",
4
+ "transformers": "4.57.0",
5
+ "pytorch": "2.7.1+cu126"
6
+ },
7
+ "model_type": "SentenceTransformer",
8
+ "prompts": {
9
+ "query": "",
10
+ "document": ""
11
+ },
12
+ "default_prompt_name": null,
13
+ "similarity_fn_name": "cosine"
14
+ }
last-checkpoint/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe8aa3a1a1fb9867e7ebcb2575f994045f392075d8e27d85f0725089e178d958
3
+ size 45437864
last-checkpoint/modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ },
14
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
last-checkpoint/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46b7d40b2fc8fc1d615cf6d14f4591edcb49cf5723c25f2421e4f1170d3494af
3
+ size 90346763
last-checkpoint/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acd6682c2d8fc2044ee522f102fcd2876d77278a945d452b98005081e2da9986
3
+ size 14645
last-checkpoint/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:198961584cd0530252f9019204a7eacfefb2ec28aecae73b687db7cd3b9da2f0
3
+ size 1465
last-checkpoint/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
last-checkpoint/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
last-checkpoint/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
last-checkpoint/tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 128,
51
+ "model_max_length": 256,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 0,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "longest_first",
64
+ "unk_token": "[UNK]"
65
+ }
last-checkpoint/trainer_state.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.8333333333333334,
6
+ "eval_steps": 500,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.20833333333333334,
14
+ "grad_norm": 2.1875,
15
+ "learning_rate": 4.4212962962962966e-05,
16
+ "loss": 1.1163,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.4166666666666667,
21
+ "grad_norm": 2.0625,
22
+ "learning_rate": 3.263888888888889e-05,
23
+ "loss": 0.6972,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.625,
28
+ "grad_norm": 2.140625,
29
+ "learning_rate": 2.1064814814814816e-05,
30
+ "loss": 0.6743,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.8333333333333334,
35
+ "grad_norm": 2.15625,
36
+ "learning_rate": 9.490740740740741e-06,
37
+ "loss": 0.6641,
38
+ "step": 200
39
+ }
40
+ ],
41
+ "logging_steps": 50,
42
+ "max_steps": 240,
43
+ "num_input_tokens_seen": 0,
44
+ "num_train_epochs": 1,
45
+ "save_steps": 200,
46
+ "stateful_callbacks": {
47
+ "TrainerControl": {
48
+ "args": {
49
+ "should_epoch_stop": false,
50
+ "should_evaluate": false,
51
+ "should_log": false,
52
+ "should_save": true,
53
+ "should_training_stop": false
54
+ },
55
+ "attributes": {}
56
+ }
57
+ },
58
+ "total_flos": 0.0,
59
+ "train_batch_size": 512,
60
+ "trial_name": null,
61
+ "trial_params": null
62
+ }
last-checkpoint/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cf87cce1a6a21832f5ea715891e737dd97976e9d192696195491423a0b6cb61
3
+ size 6161
last-checkpoint/vocab.txt ADDED
The diff for this file is too large to render. See raw diff