RikoteMaster commited on
Commit
679b553
·
verified ·
1 Parent(s): d5d59f0

Model save

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
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
+ }
README.md ADDED
@@ -0,0 +1,784 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - generated_from_trainer
7
+ - dataset_size:34441
8
+ - loss:MultipleNegativesRankingLoss
9
+ base_model: ibm-granite/granite-embedding-107m-multilingual
10
+ widget:
11
+ - source_sentence: inhibitors antiviral, antibacterial is as in with and Tzds. not
12
+ been combination with insulin. sitagliptin resulted an HbA of effects rate of
13
+ (upper tract and urinary (when (when combined with sulfonylurea), hypersensitivity
14
+ facial administered insulin agogue insulin may to be lowered to hypoglycemia.
15
+ is and to properties to sitagliptin tin. is use as with glimepiride, pioglitazone.
16
+ COMBINATION THERAPY—ORAL AGENTS & MEDICATION in Type 2 Mellitus Failure maintain
17
+ good over owing to a decrease beta-cell physical lean or increase ectopic the
18
+ manage- of type diabetes. required to glycemic Unless is a contraindication, be
19
+ initiated with a biguanide. If metformin agent or is added. drug be secret- incretin-based
20
+ therapy, amylin glucosi- given to sulfonylureas or insulin cost, include metformin,
21
+ other oral a noninsulin injectable and intensified insulin
22
+ sentences:
23
+ - inhibitors such as antiviral, antifungal, and certain antibacterial agents. Saxagliptin
24
+ is approved as monotherapy and in combination with biguanides, sulfonylureas,
25
+ and Tzds. It has not been studied in combination with insulin. During clinical
26
+ trials, mono- and combination therapy with sitagliptin resulted in an HbA 1c reduc-
27
+ tion in the range of 0.4–0.9%. Adverse effects include an increased rate of infections
28
+ (upper respiratory tract and urinary tract), headaches, peripheral edema (when
29
+ combined with a Tzd), hypoglycemia (when combined with a sulfonylurea), and hypersensitivity
30
+ reactions (urticaria, facial edema). The dose of a concurrently administered insulin
31
+ secret- agogue or insulin may need to be lowered to prevent hypoglycemia. Linagliptin
32
+ is the most recently introduced drug in this class and appears to have properties
33
+ similar to sitagliptin and saxaglip- tin. It is approved for use as monotherapy
34
+ and in combination with metformin, glimepiride, and pioglitazone. COMBINATION
35
+ THERAPY—ORAL ANTIDIABETIC AGENTS & INJECTABLE MEDICATION Combination Therapy in
36
+ Type 2 Diabetes Mellitus Failure to maintain a good response to therapy over the
37
+ long term owing to a progressive decrease in beta-cell mass, reduction in physical
38
+ activity, decline in lean body mass, or increase in ectopic fat deposition remains
39
+ a disconcerting problem in the manage- ment of type 2 diabetes. Multiple medications
40
+ may be required to achieve glycemic control. Unless there is a contraindication,
41
+ medical therapy should be initiated with a biguanide. If clinical failure occurs
42
+ with metformin monotherapy, a second agent or insulin is added. The second-line
43
+ drug can be an insulin secret- agogue, Tzd, incretin-based therapy, amylin analog,
44
+ or a glucosi- dase inhibitor; preference is given to sulfonylureas or insulin
45
+ because of cost, adverse effects, and safety concerns. Third-line therapy can
46
+ include metformin, multiple other oral medications, or a noninsulin injectable
47
+ and metformin and intensified insulin
48
+ - 'of the integrity of membranes in cells and organelles. A. Nervous System The
49
+ developing central nervous system of the fetus and young child is the most sensitive
50
+ target organ for lead’s toxic effect. Epidemiologic studies suggest that blood
51
+ lead concentrations even less than 5 mcg/dL may result in subclinical deficits
52
+ in neurocog- nitive function in lead-exposed young children, with no demon- strable
53
+ threshold for a “no effect” level. The dose response between TABLE 57–1 Toxicology
54
+ of selected arsenic, lead, and mercury compounds. Form Entering Body Major Route
55
+ of Absorption Distribution Major Clinical Effects Key Aspects of Mechanism Metabolism
56
+ and Elimination Arsenic Inorganic arsenic salts Gastrointestinal, respiratory
57
+ (all mucosal surfaces) Predominantly soft tissues (highest in liver, kidney).
58
+ Avidly bound in skin, hair, nails Cardiovascular: shock, arrhythmias. CNS: encephalopathy,
59
+ peripheral neuropathy. Gastroenteritis; pan- cytopenia; cancer (many sites) Inhibits
60
+ enzymes; interferes with oxidative phosphorylation; alters cell signaling, gene
61
+ expression Methylation. Renal (major); sweat and feces (minor) Lead Inorganic
62
+ lead oxides and salts Gastrointestinal, respiratory Soft tissues; redistributed
63
+ to skeleton (> 90% of adult body burden) CNS deficits; peripheral neuropathy;
64
+ ane- mia; nephropathy; hypertension; reproductive toxicity Inhibits enzymes; interferes
65
+ with essential cations; alters membrane structure Renal (major); feces and breast
66
+ milk (minor) Organic (tetraethyl lead) Skin, gastrointesti- nal, respiratory Soft
67
+ tissues, especially liver, CNS Encephalopathy Hepatic dealkylation (fast) → trialkyme-
68
+ tabolites (slow) → dissociation to lead Urine and feces (major); sweat (minor)
69
+ Mercury Elemental mercury Respiratory tract Soft tissues, especially kidney, CNS
70
+ CNS: tremor, behavioral (erethism); gingivo'
71
+ - '61. Glucocorticoids for gastrointestinal use: See Chapter 62. REFERENCES Alesci
72
+ S et al: Glucocorticoid-induced osteoporosis: From basic mechanisms to clinical
73
+ aspects. Neuroimmunomodulation 2005;12:1. Bamberger CM, Schulte HM, Chrousos GP:
74
+ Molecular determinants of gluco- corticoid receptor function and tissue sensitivity
75
+ to glucocorticoids. Endocr Rev 1996;17:245. Charmandari E, Kino T: Chrousos syndrome:
76
+ A seminal report, a phylogenetic enigma and the clinical implications of glucocorticoid
77
+ signaling changes. Eur J Clin Invest 2010;40:932. Charmandari E, Tsigos C, Chrousos
78
+ GP: Neuroendocrinology of stress. Ann Rev Physiol 2005;67:259. Chrousos GP: Stress
79
+ and disorders of the stress system. Nat Endocrinol Rev 2009;5:374. Chrousos GP,
80
+ Kino T: Glucocorticoid signaling in the cell: Expanding clinical implications
81
+ to complex human behavioral and somatic disorders. In: Glucocorticoids and mood:
82
+ Clinical manifestations, risk factors, and molecular mechanisms. Proc NY Acad
83
+ Sci 2009;1179:153. Elenkov IJ, Chrousos GP: Stress hormones, TH1/TH2 patterns,
84
+ pro/anti-in- flammatory cytokines and susceptibility to disease. Trends Endocrinol
85
+ Metab 1999;10:359. Elenkov IJ et al: Cytokine dysregulation, inflammation, and
86
+ wellbeing. Neuroimmunomodulation 2005;12:255. Franchimont D et al: Glucocorticoids
87
+ and inflammation revisited: The state of the art. Neuroimmunomodulation 2002–03;10:247.
88
+ Graber AL et al: Natural history of pituitary-adrenal recovery following long-term
89
+ suppression with corticosteroids. J Clin Endocrinol Metab 1965;25:11. Hochberg
90
+ Z, Pacak K, Chrousos GP: Endocrine withdrawal syndromes. Endocrine Rev 2003;24:523.
91
+ Kalantaridou S, Chrousos GP: Clinical review 148:'
92
+ - source_sentence: Against Gram-Negative Carbapenems Cephalosporins Chloramphenicol
93
+ Daptomycin Tigecycline Oxazolidinones Penicillins Streptogramins Trimethoprim
94
+ TABLE 51–5 agents that dosage or are contraindicated patients hepatic Dosage Adjustment
95
+ in Impairment Renal Impairment Dosage Needed Hepatic Impairment Acyclovir, aztreonam,
96
+ carbapenems, clarithromycin, colistin, cycloserine, daptomycin, didanosine, ethionamide,
97
+ famciclovir, foscarnet, ganciclovir, penicillins,3 stavudine, telithromycin, tenofovir,
98
+ trimethoprim- Cidofovir, tetracyclines2 Amprenavir, atazanavir, erythromycin,
99
+ 1Except 2Except doxycycline minocycline. 3Except antistaphylococcal penicillins
100
+ (eg, dicloxacillin). That Alter Antimicrobi
101
+ sentences:
102
+ - the body to colonize various organs in the process called metastasis. Such tumor
103
+ stem cells thus can express clonogenic (colony-forming) capability, and they are
104
+ characterized by chromosome abnormalities reflecting their genetic instability,
105
+ which leads to progressive selection of subclones that can survive more readily
106
+ in the multicellular environment of the host. This genetic instability also allows
107
+ them to become resistant to chemotherapy and radiotherapy. The invasive and metastatic
108
+ processes as well as a series of metabolic abnormalities associated with the cancer
109
+ result in tumor-related symptoms and eventual death of the patient unless the
110
+ neoplasm can be eradicated with treatment. 54 CAUSES OF CANCER The incidence,
111
+ geographic distribution, and behavior of specific types of cancer are related
112
+ to multiple factors, including sex, age, race, genetic predisposition, and exposure
113
+ to environmental car- cinogens. Of these factors, environmental exposure is probably
114
+ most important. Exposure to ionizing radiation has been well documented as a significant
115
+ risk factor for a number of cancers, including acute leukemias, thyroid cancer,
116
+ breast cancer, lung cancer, soft tissue sarcoma, and basal cell and squamous cell
117
+ skin cancers. Chemical carcinogens (particularly those in tobacco smoke) as well
118
+ as azo dyes, aflatoxins, asbestos, benzene, and radon have all been well documented
119
+ as leading to a wide range of human cancers. Several viruses have been implicated
120
+ in the etiology of various human cancers. For example, hepatitis B and hepatitis
121
+ C are asso- ciated with the development of hepatocellular cancer; HIV is associated
122
+ with Hodgkin’s and non-Hodgkin’s lymphomas; human papillomavirus is associated
123
+ with cervical cancer and head and neck cancer; and Ebstein-Barr virus is associated
124
+ with nasopharyn- geal cancer. Expression of virus-induced neoplasia may also depend
125
+ on additional host and environmental factors that modu- late the transformation
126
+ process. Cellular genes are known that are homologous to the transforming genes
127
+ of the retroviruses, a family
128
+ - Against Gram-Positive Cocci Against Gram-Negative Bacilli Aminoglycosides Aminoglycosides
129
+ Carbapenems Carbapenems Cephalosporins Chloramphenicol Chloramphenicol Quinolones
130
+ Clindamycin Rifampin Daptomycin Tetracyclines Glycopeptide antibiotics Tigecycline
131
+ Ketolides Macrolides Oxazolidinones Penicillins Quinolones Rifampin Streptogramins
132
+ Sulfonamides Tetracyclines Tigecycline Trimethoprim TABLE 51–5 Antimicrobial agents
133
+ that require dosage adjustment or are contraindicated in patients with renal or
134
+ hepatic impairment. Dosage Adjustment Needed in Renal Impairment Contraindicated
135
+ in Renal Impairment Dosage Adjustment Needed in Hepatic Impairment Acyclovir,
136
+ amantadine, aminoglycosides, aztreonam, carbapenems, cephalosporins,1 clarithromycin,
137
+ colistin, cycloserine, daptomycin, didanosine, emtricitabine, ethambutol, ethionamide,
138
+ famciclovir, fluconazole, flucytosine, foscarnet, ganciclovir, lamivudine, penicillins,3
139
+ pyrazinamide, quinolones, 4 rimantadine, stavudine, telavancin, telbivudine, telithromycin,
140
+ tenofovir, terbinafine, trimethoprim- sulfamethoxazole, valacyclovir, vancomycin,
141
+ zidovudine Cidofovir, methenamine, nalidixic acid, nitrofurantoin, sulfonamides
142
+ (long-acting), tetracyclines2 Amprenavir, atazanavir, chloram- phenicol, clindamycin,
143
+ erythromycin, fosamprenavir, indinavir, metronida- zole, rimantadine, tigecycline
144
+ 1Except ceftriaxone. 2Except doxycycline and possibly minocycline. 3Except antistaphylococcal
145
+ penicillins (eg, nafcillin and dicloxacillin). 4Except moxifloxacin. Conditions
146
+ That Alter Antimicrobi
147
+ - host disease after allogeneic stem cell trans- plantation. Cyclosporine has also
148
+ proved useful in a variety of autoimmune disorders, including uveitis, rheumatoid
149
+ arthritis, psoriasis, and asthma. Its combination with newer agents is show- ing
150
+ considerable efficacy in clinical and experimental settings where effective and
151
+ less toxic immunosuppression is needed. Newer for- mulations of cyclosporine have
152
+ been developed that are improving patient compliance (smaller, better tasting
153
+ pills) and increasing bioavailability. Tacrolimus Tacrolimus (FK 506) is an immunosuppressant
154
+ macrolide antibi- otic produced by Streptomyces tsukubaensis. It is not chemically
155
+ related to cyclosporine, but their mechanisms of action are similar. Both drugs
156
+ bind to cytoplasmic peptidylprolyl isomerases that are abundant in all tissues.
157
+ While cyclosporine binds to cyclophilin, tacrolimus binds to the immunophilin
158
+ FK-binding protein (FKBP). Both complexes inhibit calcineurin, which is necessary
159
+ for the activation of the T-cell-specific transcription factor NF-AT. On a weight
160
+ basis, tacrolimus is 10–100 times more potent than cyclosporine in inhibiting
161
+ immune responses. Tacrolimus is utilized for the same indications as cyclosporine,
162
+ particularly in organ and stem cell transplantation. Multicenter studies in the
163
+ USA and in Europe indicate that both graft and patient survival are similar for
164
+ the two drugs. Tacrolimus has proved to be effective therapy for preventing rejection
165
+ in solid-organ transplant patients even after failure of standard rejection therapy,
166
+ including anti-T- cell antibodies. It is now considered a standard prophylactic
167
+ agent (usually in combination with methotrexate or mycophenolate mofetil) for
168
+ graft-versus-host disease. Tacrolimus can be administered orally or intravenously.
169
+ The half-life of the intravenous form is approximately 9–12 hours. Like cyclosporine,
170
+ tacrolimus is metabolized primarily by P450 enzymes in the liver, and there is
171
+ potential for drug interactions. The dosage is determined by trough blood level
172
+ at
173
+ - source_sentence: Antiprotozoal 923 therapy many strains of P and is a of recommended
174
+ chemopro- for in malaria-endemic with chloroquine-resistant is a synthetic 4-quinoline
175
+ is related to quinine. It only be given orally because local with parenteral It
176
+ is absorbed, and peak reached 18 Mefloquine is extensively distrib- uted and eliminated
177
+ treat- ment The terminal elimination about 20 days, weekly dosing ing, steady-state
178
+ levels are over a weeks; this shortened beginning a with consecutive although
179
+ is stan- practice. slowly mainly the The can in the after of & Mefloquine has
180
+ P falciparum vivax, but not against The mechanism of action is to mefloquine from
181
+ areas. appears regions Mefloquine quinine with Clinical Uses A. Chemoprophylaxis
182
+ Mefloquine prophylaxis most
183
+ sentences:
184
+ - CHAPTER 52 Antiprotozoal Drugs 923 MEFLOQUINE Mefloquine is effective therapy
185
+ for many chloroquine-resistant strains of P falciparum and against other species.
186
+ Although toxicity is a concern, mefloquine is one of the recommended chemopro-
187
+ phylactic drugs for use in most malaria-endemic regions with chloroquine-resistant
188
+ strains. Chemistry & Pharmacokinetics Mefloquine hydrochloride is a synthetic
189
+ 4-quinoline methanol that is chemically related to quinine. It can only be given
190
+ orally because severe local irritation occurs with parenteral use. It is well
191
+ absorbed, and peak plasma concentrations are reached in about 18 hours. Mefloquine
192
+ is highly protein-bound, extensively distrib- uted in tissues, and eliminated
193
+ slowly, allowing a single-dose treat- ment regimen. The terminal elimination half-life
194
+ is about 20 days, allowing weekly dosing for chemoprophylaxis. With weekly dos-
195
+ ing, steady-state drug levels are reached over a number of weeks; this interval
196
+ can be shortened to 4 days by beginning a course with three consecutive daily
197
+ doses of 250 mg, although this is not stan- dard practice. Mefloquine and acid
198
+ metabolites of the drug are slowly excreted, mainly in the feces. The drug can
199
+ be detected in the blood for months after the completion of therapy. Antimalarial
200
+ Action & Resistance Mefloquine has strong blood schizonticidal activity against
201
+ P falciparum and P vivax, but it is not active against hepatic stages or gametocytes.
202
+ The mechanism of action of mefloquine is unknown. Sporadic resistance to mefloquine
203
+ has been reported from many areas. At present, resistance appears to be uncommon
204
+ except in regions of Southeast Asia with high rates of multidrug resistance (especially
205
+ border areas of Thailand). Mefloquine resis- tance appears to be associated with
206
+ resistance to quinine and halofantrine but not with resistance to chloroquine.
207
+ Clinical Uses A. Chemoprophylaxis Mefloquine is effective in prophylaxis against
208
+ most strain
209
+ - 938 SECTION VIII Chemotherapeutic Drugs Clinical Uses Albendazole is administered
210
+ on an empty stomach when used against intraluminal parasites but with a fatty
211
+ meal when used against tissue parasites. A. Ascariasis, Trichuriasis, and Hookworm
212
+ and Pinworm Infections For adults and children older than 2 years of age with
213
+ ascariasis and hookworm infections, the treatment is a single dose of 400 mg TABLE
214
+ 53–1 Drugs for the treatment of helminthic infections. 1 Infecting Organism Drug
215
+ of Choice Alternative Drugs Roundworms (nematodes) Ascaris lumbricoides (roundworm)
216
+ Albendazole or pyrantel pamoate or mebendazole Ivermectin, piperazine Trichuris
217
+ trichiura (whipworm) Mebendazole or albendazole Ivermectin Necator americanus
218
+ (hookworm); Ancylostoma duodenale (hookworm) Albendazole or mebendazole or pyrantel
219
+ pamoate Strongyloides stercoralis (threadworm) Ivermectin Albendazole or thiabendazole
220
+ Enterobius vermicularis (pinworm) Mebendazole or pyrantel pamoate Albendazole
221
+ Trichinella spiralis (trichinosis) Mebendazole or albendazole; add corticosteroids
222
+ for severe infection Trichostrongylus species Pyrantel pamoate or mebendazole
223
+ Albendazole Cutaneous larva migrans (creeping eruption) Albendazole or ivermectin
224
+ Thiabendazole (topical) Visceral larva migrans Albendazole Mebendazole Angiostrongylus
225
+ cantonensis Albendazole or mebendazole Wuchereria bancrofti (filariasis); Brugia
226
+ malayi (filariasis); tropical eosinophilia; Loa loa (loiasis) Diethylcarbamazine
227
+ Ivermectin Onchocerca volvulus (onchocerciasis) Ivermectin Dracunculus medinensis
228
+ (guinea worm) Metronidazole Thiabendazole or mebendazole Capillaria philippinensis
229
+ (intestinal capillariasis) Albendazole Mebendazole Flukes (trematodes) Schistosoma
230
+ haematobium (bilharziasis)
231
+ - safely and effectively combined with 5-FU-, irinotecan-, and oxaliplatin-based
232
+ chemotherapy in the treatment of metastatic colorectal cancer. Bevacizumab is
233
+ FDA approved as a first-line treatment for metastatic colorectal cancer in combination
234
+ with any intravenous fluoropyrimidine-contain- ing regimen and is now also approved
235
+ in combination with che- motherapy for metastatic non-small lung cancer and breast
236
+ cancer. One potential advantage of this antibody is that it does not appear to
237
+ exacerbate the toxicities typically observed with cytotoxic che- motherapy. The
238
+ main safety concerns associated with bevacizumab include hypertension, an increased
239
+ incidence of arterial throm- boembolic events (transient ischemic attack, stroke,
240
+ angina, and myocardial infarction), wound healing complications, gastrointes-
241
+ tinal perforations, and proteinuria. Sorafenib is a small molecule that inhibits
242
+ multiple receptor tyrosine kinases (RTKs), especially VEGF-R2 and VEGF-R3, platelet-derived
243
+ growth factor-β (PDGFR-β), and raf kinase. It was initially approved for advanced
244
+ renal cell cancer and is also approved for advanced hepatocellular cancer. Sunitinib
245
+ is similar to sorafenib in that it inhibits multiple RTKs, although the specific
246
+ types are somewhat different. They include PDGFR-α and PDGFR-β, VEGF-R1, VEGF-R2,
247
+ VEGF-R3, and c-kit. It is approved for the treatment of advanced renal cell cancer
248
+ and for the treatment of gastrointestinal stromal tumors (GIST) after disease
249
+ progression on or with intolerance to imatinib. Pazopanib is a small molecule
250
+ that inhibits multiple RTKs, espe- cially VEGF-R2 and VEGF-R3, PDGFR-β, and raf
251
+ kinase. This oral agent is approved for the treatment of advanced renal cell cancer.
252
+ Sorafenib, sunitinib, and pazopanib are metabolized in the liver by the CYP3A4
253
+ system, and elimination is primarily hepatic with excretion in feces. Each of
254
+ these agents has potential interac-
255
+ - source_sentence: Endocrine is gland, increases phate reduce the enhanced feedback
256
+ regulation the effect PTH to calcium and reduce Likewise, and at levels the D
257
+ kidney increase amount produced. High reducing PTH works by FGF23 1,25(OH) 2 raises
258
+ phosphate, whereas 2 D has such is appropriate. 1,25(OH) D of on serum inhibitory
259
+ effect negative feedback patients producing 1,25(OH) loss feedback coupled with
260
+ impaired and intestinal calcium leads to hyperparathyroidism. The of 2 D inhibit
261
+ being exploited with analogs serum calcium their drugs are useful roidism accompanying
262
+ chronic disease in of primary 1,25(OH) also stimulates production completes negative
263
+ feedback loop inhibits 1,25(OH) production while promoting hypophosphatemia, which
264
+ turn production 1,25(OH) D production. SECONDARY HOMEOST
265
+ sentences:
266
+ - 774 SECTION VII Endocrine Drugs that is detected by the parathyroid gland, increases
267
+ in serum phos- phate levels reduce the ionized calcium, leading to enhanced PTH
268
+ secretion. Such feedback regulation is appropriate to the net effect of PTH to
269
+ raise serum calcium and reduce serum phosphate levels. Likewise, both calcium
270
+ and phosphate at high levels reduce the amount of 1,25(OH) 2 D produced by the
271
+ kidney and increase the amount of 24,25(OH) 2 D produced. High serum calcium works
272
+ directly and indirectly by reducing PTH secretion. High serum phosphate works
273
+ directly and indirectly by increasing FGF23 levels. Since 1,25(OH) 2 D raises
274
+ serum calcium and phosphate, whereas 24,25(OH) 2 D has less effect, such feedback
275
+ regulation is again appropriate. 1,25(OH) 2 D directly inhibits PTH secretion
276
+ (independent of its effect on serum calcium) by a direct inhibitory effect on
277
+ PTH gene transcription. This pro- vides yet another negative feedback loop. In
278
+ patients with chronic renal failure who frequently are deficient in producing
279
+ 1,25(OH) 2 D, loss of this 1,25(OH) 2 D-mediated feedback loop coupled with impaired
280
+ phosphate excretion and intestinal calcium absorption often leads to secondary
281
+ hyperparathyroidism. The ability of 1,25(OH) 2 D to inhibit PTH secretion directly
282
+ is being exploited with calcitriol analogs that have less effect on serum calcium
283
+ because of their lesser effect on intestinal calcium absorption. Such drugs are
284
+ proving useful in the management of secondary hyperparathy- roidism accompanying
285
+ chronic kidney disease and may be useful in selected cases of primary hyperparathyroidism.
286
+ 1,25(OH) 2 D also stimulates the production of FGF23. This completes the negative
287
+ feedback loop in that FGF23 inhibits 1,25(OH) 2 D production while promoting hypophosphatemia,
288
+ which in turn inhibits FGF23 production and stimulates 1,25(OH) 2 D production.
289
+ SECONDARY HORMONAL REGULATORS OF BONE MINERAL HOMEOST
290
+ - ke). Equine and ovine antivenoms are available for rattle- snake envenomations,
291
+ but only equine antivenom is available for coral snake bite. The ovine antivenom
292
+ is a Fab preparation and is less immunogenic than whole equine IgG antivenoms,
293
+ but retains the ability to neutralize the rattlesnake venom. MONOCLONAL ANTIBODIES
294
+ (MABs ) Recent advances in the ability to manipulate the genes of immu- noglobulins
295
+ have resulted in development of a wide array of humanized and chimeric monoclonal
296
+ antibodies directed against therapeutic targets. The only murine elements of humanized
297
+ monoclonal antibodies are the complementarity-determining regions in the variable
298
+ domains of immunoglobulin heavy and light chains. Complementarity-determining
299
+ regions are primarily responsible for the antigen-binding capacity of antibodies.
300
+ Chimeric antibodies typically contain antigen-binding murine variable regions
301
+ and human constant regions. The following are brief descriptions of the engineered
302
+ antibodies that have been approved by the FDA. Antitumor MABs Alemtuzumab is a
303
+ humanized IgG 1 with a kappa chain that binds to CD52 found on normal and malignant
304
+ B and T lymphocytes, NK cells, monocytes, macrophages, and a small population
305
+ of granulocytes. Currently, alemtuzumab is approved for the treatment of B-cell
306
+ chronic lymphocytic leukemia in patients who have been treated with alkylating
307
+ agents and have failed fludarabine therapy. Alemtuzumab appears to deplete leukemic
308
+ and normal cells by direct antibody-dependent lysis. Patients receiving this antibody
309
+ become lymphopenic and may also become neutro- penic, anemic, and thrombocytopenic.
310
+ As a result patients should be closely monitored for opportunistic infections
311
+ and hemato- logic toxicity. Bevacizumab is a humanized IgG 1 monoclonal antibody
312
+ that binds to vascular endothelial growth factor (VEGF) and inhibits VEGF from
313
+ binding to its receptor, especially on endothelial cells. It is an antiangiogenic
314
+ drug that
315
+ - rier only when the meninges are inflamed. Concentrations in cerebrospinal fluid
316
+ are highly variable, ranging from 4% to 64% of serum levels in the setting of
317
+ meningeal inflammation. As with all antituberculous drugs, resistance to ethambutol
318
+ emerges rapidly when the drug is used alone. Therefore, ethambutol is always given
319
+ in combination with other antituberculous drugs. Ethambutol hydrochloride, 15–25
320
+ mg/kg, is usually given as a single daily dose in combination with isoniazid or
321
+ rifampin. The higher dose is recommended for treatment of tuberculous menin- gitis.
322
+ The dose of ethambutol is 50 mg/kg when a twice-weekly dosing schedule is used.
323
+ Adverse Reactions Hypersensitivity to ethambutol is rare. The most common serious
324
+ adverse event is retrobulbar neuritis, resulting in loss of visual acuity and
325
+ red-green color blindness. This dose-related adverse effect is more likely to
326
+ occur at dosages of 25 mg/kg/d continued for several months. At 15 mg/kg/d or
327
+ less, visual disturbances are very rare. Periodic visual acuity testing is desirable
328
+ if the 25 mg/kg/d dosage is used. Ethambutol is relatively contraindicated in
329
+ chil- dren too young to permit assessment of visual acuity and red- green color
330
+ discrimination. PYRAZINAMIDE Pyrazinamide (PZA) is a relative of nicotinamide.
331
+ It is stable and slightly soluble in water. It is inactive at neutral pH, but
332
+ at pH 5.5 it inhibits tubercle bacilli at concentrations of approximately 20 mcg/mL.
333
+ The drug is taken up by macrophages and exerts its activity against mycobacteria
334
+ residing within the acidic environ- ment of lysosomes. Pyrazinamide (PZA) N C
335
+ O NH2 N Mechanism of Action & Clinical Uses Pyrazinamide is converted to pyrazinoic
336
+ acid—the active form of the drug—by mycobacterial pyrazinamidase, which is encoded
337
+ by
338
+ - source_sentence: Agents 49–1 Agents to or prevent herpes simplex virus and varicella-zoster
339
+ virus (VZV) Route Administration Use Recommended Dosage and Regimen Acyclovir1
340
+ First herpes 400 tid 5 times daily 7–10 days Recurrent genital herpes treatment
341
+ or 200 daily or 800 bid or 800 mg tid × days in the HIV-infected mg 3–5 daily
342
+ days in the HIV-infected mg times Orolabial herpes 400 mg 5 × 5 treatment 800
343
+ qid treatment mg 5 days Intravenous mg/kg Mucocutaneous herpes the host treatment
344
+ q8h × mg/kg × days HSV 10–20 Varicella or zoster in the immunosuppressed host
345
+ × Topical (5% cream) Herpes labialis covering times × 4 days First genital herpes
346
+ 500 × Recurrent treatment × 1 Genital in bid Genital herpes herpes suppression
347
+ the mg bid 1500 mg Orolabial suppression 250-500 mg mg tid × 7 treatment 10 days
348
+ Recurrent treatment 500 days Genital in HIV-infected treatment 5–10 herpes herpes
349
+ suppression HIV
350
+ sentences:
351
+ - Primaquine is the drug of choice for the eradication of dormant liver forms of
352
+ P vivax and P ovale and can also be used for chemo- prophylaxis against all malarial
353
+ species. Chemistry & Pharmacokinetics Primaquine phosphate is a synthetic 8-aminoquinoline
354
+ ( Figure 52–2 ). The drug is well absorbed orally, reaching peak plasma levels
355
+ in
356
+ - CHAPTER 49 Antiviral Agents 865 TABLE 49–1 Agents to treat or prevent herpes simplex
357
+ virus (HSV) and varicella-zoster virus (VZV) infections. Route of Administration
358
+ Use Recommended Adult Dosage and Regimen Acyclovir1 Oral First episode genital
359
+ herpes treatment 400 mg tid or 200 mg 5 times daily × 7–10 days Recurrent genital
360
+ herpes treatment 400 mg tid or 200 mg 5 times daily or 800 mg bid × 3–5 days or
361
+ 800 mg tid × 2 days Genital herpes in the HIV-infected host treatment 400 mg 3–5
362
+ times daily × 5–10 days Genital herpes suppression in the HIV-infected host 400–800
363
+ mg bid–tid Herpes proctitis treatment 400 mg 5 times daily until healed Orolabial
364
+ herpes treatment 400 mg 5 times daily × 5 days Varicella treatment (age ≥ 2 years)
365
+ 800 mg qid × 5 days Zoster treatment 800 mg 5 times daily × 7–10 days Intravenous
366
+ Severe HSV treatment 5 mg/kg q8h × 7–10 days Mucocutaneous herpes in the immunocompromised
367
+ host treatment 10 mg/kg q8h × 7–14 days Herpes encephalitis treatment 10–15 mg/kg
368
+ q8h × 14–21 days Neonatal HSV infection treatment 10–20 mg/kg q8h × 14–21 days
369
+ Varicella or zoster in the immunosuppressed host treatment 10 mg/kg q8h × 7 days
370
+ Topical (5% cream) Herpes labialis treatment Thin film covering lesion 5 times
371
+ daily × 4 days Famciclovir1 Oral First episode genital herpes treatment 500 mg
372
+ tid × 5–10 days Recurrent genital herpes treatment 1000 mg bid × 1 day Genital
373
+ herpes in the HIV-infected host treatment 500 mg bid × 5–10 days Genital herpes
374
+ suppression 250 mg bid Genital herpes suppression in the HIV-infected host 500
375
+ mg bid Orolabial herpes treatment 1500 mg once Orolabial or genital herpes suppression
376
+ 250-500 mg bid Zoster 500 mg tid × 7 days Valacyclovir1 Oral First episode genital
377
+ herpes treatment 1000 mg bid × 10 days Recurrent genital herpes treatment 500
378
+ mg bid × 3 days Genital herpes in the HIV-infected host treatment 500–1000 mg
379
+ bid × 5–10 days Genital herpes suppression 500–1000 mg once daily Genital herpes
380
+ suppression in the HIV
381
+ - 708 SECTION VII Endocrine Drugs marked adverse effects because there is a recovery
382
+ period between each dose. The transition to an alternate-day schedule can be made
383
+ after the disease process is under control. It should be done gradu- ally and
384
+ with additional supportive measures between doses. When selecting a drug for use
385
+ in large doses, a medium- or intermediate-acting synthetic steroid with little
386
+ mineralocorticoid effect is advisable. If possible, it should be given as a single
387
+ morning dose. C. Special Dosage Forms Local therapy, such as topical preparations
388
+ for skin disease, oph- thalmic forms for eye disease, intra-articular injections
389
+ for joint disease, inhaled steroids for asthma, and hydrocortisone enemas for
390
+ ulcerative colitis, provides a means of delivering large amounts of steroid to
391
+ the diseased tissue with reduced systemic effects. Beclomethasone dipropionate,
392
+ and several other glucocorti- coids—primarily budesonide, flunisolide, and mometasone
393
+ furoate, administered as aerosols—have been found to be extremely useful in the
394
+ treatment of asthma (see Chapter 20 ). Beclomethasone dipropionate, triamcinolone
395
+ acetonide, budes- onide, flunisolide, and mometasone furoate are available as
396
+ nasal sprays for the topical treatment of allergic rhinitis. They are effec- tive
397
+ at doses (one or two sprays one, two, or three times daily) that in most patients
398
+ result in plasma levels that are too low to influ- ence adrenal function or have
399
+ any other systemic effects. Corticosteroids incorporated in ointments, creams,
400
+ lotions, and sprays are used extensively in dermatology. These preparations are
401
+ discussed in more detail in Chapter 61 . MINERALOCORTICOIDS (ALDOSTERONE, DEOXYCORTICOSTERONE,
402
+ FLUDROCORTISONE) The most important mineralocorticoid in humans is aldosterone.
403
+ However, small amounts of deoxycorticosterone (DOC) are also formed and released.
404
+ Although the amount is normally insignifi- cant, DOC was of some importance therapeut
405
+ pipeline_tag: sentence-similarity
406
+ library_name: sentence-transformers
407
+ ---
408
+
409
+ # SentenceTransformer based on ibm-granite/granite-embedding-107m-multilingual
410
+
411
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [ibm-granite/granite-embedding-107m-multilingual](https://huggingface.co/ibm-granite/granite-embedding-107m-multilingual). 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.
412
+
413
+ ## Model Details
414
+
415
+ ### Model Description
416
+ - **Model Type:** Sentence Transformer
417
+ - **Base model:** [ibm-granite/granite-embedding-107m-multilingual](https://huggingface.co/ibm-granite/granite-embedding-107m-multilingual) <!-- at revision 5c793ec061753b0d0816865e1af7db3f675d65af -->
418
+ - **Maximum Sequence Length:** 512 tokens
419
+ - **Output Dimensionality:** 384 dimensions
420
+ - **Similarity Function:** Cosine Similarity
421
+ <!-- - **Training Dataset:** Unknown -->
422
+ <!-- - **Language:** Unknown -->
423
+ <!-- - **License:** Unknown -->
424
+
425
+ ### Model Sources
426
+
427
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
428
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
429
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
430
+
431
+ ### Full Model Architecture
432
+
433
+ ```
434
+ SentenceTransformer(
435
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
436
+ (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
437
+ (2): Normalize()
438
+ )
439
+ ```
440
+
441
+ ## Usage
442
+
443
+ ### Direct Usage (Sentence Transformers)
444
+
445
+ First install the Sentence Transformers library:
446
+
447
+ ```bash
448
+ pip install -U sentence-transformers
449
+ ```
450
+
451
+ Then you can load this model and run inference.
452
+ ```python
453
+ from sentence_transformers import SentenceTransformer
454
+
455
+ # Download from the 🤗 Hub
456
+ model = SentenceTransformer("RikoteMaster/embedder-granite")
457
+ # Run inference
458
+ sentences = [
459
+ 'Agents 49–1 Agents to or prevent herpes simplex virus and varicella-zoster virus (VZV) Route Administration Use Recommended Dosage and Regimen Acyclovir1 First herpes 400 tid 5 times daily 7–10 days Recurrent genital herpes treatment or 200 daily or 800 bid or 800 mg tid × days in the HIV-infected mg 3–5 daily days in the HIV-infected mg times Orolabial herpes 400 mg 5 × 5 treatment 800 qid treatment mg 5 days Intravenous mg/kg Mucocutaneous herpes the host treatment q8h × mg/kg × days HSV 10–20 Varicella or zoster in the immunosuppressed host × Topical (5% cream) Herpes labialis covering times × 4 days First genital herpes 500 × Recurrent treatment × 1 Genital in bid Genital herpes herpes suppression the mg bid 1500 mg Orolabial suppression 250-500 mg mg tid × 7 treatment 10 days Recurrent treatment 500 days Genital in HIV-infected treatment 5–10 herpes herpes suppression HIV',
460
+ 'CHAPTER 49 Antiviral Agents 865 TABLE 49–1 Agents to treat or prevent herpes simplex virus (HSV) and varicella-zoster virus (VZV) infections. Route of Administration Use Recommended Adult Dosage and Regimen Acyclovir1 Oral First episode genital herpes treatment 400 mg tid or 200 mg 5 times daily × 7–10 days Recurrent genital herpes treatment 400 mg tid or 200 mg 5 times daily or 800 mg bid × 3–5 days or 800 mg tid × 2 days Genital herpes in the HIV-infected host treatment 400 mg 3–5 times daily × 5–10 days Genital herpes suppression in the HIV-infected host 400–800 mg bid–tid Herpes proctitis treatment 400 mg 5 times daily until healed Orolabial herpes treatment 400 mg 5 times daily × 5 days Varicella treatment (age ≥ 2 years) 800 mg qid × 5 days Zoster treatment 800 mg 5 times daily × 7–10 days Intravenous Severe HSV treatment 5 mg/kg q8h × 7–10 days Mucocutaneous herpes in the immunocompromised host treatment 10 mg/kg q8h × 7–14 days Herpes encephalitis treatment 10–15 mg/kg q8h × 14–21 days Neonatal HSV infection treatment 10–20 mg/kg q8h × 14–21 days Varicella or zoster in the immunosuppressed host treatment 10 mg/kg q8h × 7 days Topical (5% cream) Herpes labialis treatment Thin film covering lesion 5 times daily × 4 days Famciclovir1 Oral First episode genital herpes treatment 500 mg tid × 5–10 days Recurrent genital herpes treatment 1000 mg bid × 1 day Genital herpes in the HIV-infected host treatment 500 mg bid × 5–10 days Genital herpes suppression 250 mg bid Genital herpes suppression in the HIV-infected host 500 mg bid Orolabial herpes treatment 1500 mg once Orolabial or genital herpes suppression 250-500 mg bid Zoster 500 mg tid × 7 days Valacyclovir1 Oral First episode genital herpes treatment 1000 mg bid × 10 days Recurrent genital herpes treatment 500 mg bid × 3 days Genital herpes in the HIV-infected host treatment 500–1000 mg bid × 5–10 days Genital herpes suppression 500–1000 mg once daily Genital herpes suppression in the HIV',
461
+ '708 SECTION VII Endocrine Drugs marked adverse effects because there is a recovery period between each dose. The transition to an alternate-day schedule can be made after the disease process is under control. It should be done gradu- ally and with additional supportive measures between doses. When selecting a drug for use in large doses, a medium- or intermediate-acting synthetic steroid with little mineralocorticoid effect is advisable. If possible, it should be given as a single morning dose. C. Special Dosage Forms Local therapy, such as topical preparations for skin disease, oph- thalmic forms for eye disease, intra-articular injections for joint disease, inhaled steroids for asthma, and hydrocortisone enemas for ulcerative colitis, provides a means of delivering large amounts of steroid to the diseased tissue with reduced systemic effects. Beclomethasone dipropionate, and several other glucocorti- coids—primarily budesonide, flunisolide, and mometasone furoate, administered as aerosols—have been found to be extremely useful in the treatment of asthma (see Chapter 20 ). Beclomethasone dipropionate, triamcinolone acetonide, budes- onide, flunisolide, and mometasone furoate are available as nasal sprays for the topical treatment of allergic rhinitis. They are effec- tive at doses (one or two sprays one, two, or three times daily) that in most patients result in plasma levels that are too low to influ- ence adrenal function or have any other systemic effects. Corticosteroids incorporated in ointments, creams, lotions, and sprays are used extensively in dermatology. These preparations are discussed in more detail in Chapter 61 . MINERALOCORTICOIDS (ALDOSTERONE, DEOXYCORTICOSTERONE, FLUDROCORTISONE) The most important mineralocorticoid in humans is aldosterone. However, small amounts of deoxycorticosterone (DOC) are also formed and released. Although the amount is normally insignifi- cant, DOC was of some importance therapeut',
462
+ ]
463
+ embeddings = model.encode(sentences)
464
+ print(embeddings.shape)
465
+ # [3, 384]
466
+
467
+ # Get the similarity scores for the embeddings
468
+ similarities = model.similarity(embeddings, embeddings)
469
+ print(similarities.shape)
470
+ # [3, 3]
471
+ ```
472
+
473
+ <!--
474
+ ### Direct Usage (Transformers)
475
+
476
+ <details><summary>Click to see the direct usage in Transformers</summary>
477
+
478
+ </details>
479
+ -->
480
+
481
+ <!--
482
+ ### Downstream Usage (Sentence Transformers)
483
+
484
+ You can finetune this model on your own dataset.
485
+
486
+ <details><summary>Click to expand</summary>
487
+
488
+ </details>
489
+ -->
490
+
491
+ <!--
492
+ ### Out-of-Scope Use
493
+
494
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
495
+ -->
496
+
497
+ <!--
498
+ ## Bias, Risks and Limitations
499
+
500
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
501
+ -->
502
+
503
+ <!--
504
+ ### Recommendations
505
+
506
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
507
+ -->
508
+
509
+ ## Training Details
510
+
511
+ ### Training Dataset
512
+
513
+ #### Unnamed Dataset
514
+
515
+ * Size: 34,441 training samples
516
+ * Columns: <code>anchor</code> and <code>positive</code>
517
+ * Approximate statistics based on the first 1000 samples:
518
+ | | anchor | positive |
519
+ |:--------|:-----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
520
+ | type | string | string |
521
+ | details | <ul><li>min: 4 tokens</li><li>mean: 99.53 tokens</li><li>max: 279 tokens</li></ul> | <ul><li>min: 14 tokens</li><li>mean: 245.16 tokens</li><li>max: 512 tokens</li></ul> |
522
+ * Samples:
523
+ | anchor | positive |
524
+ |:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
525
+ | <code>Advanced March 2022 Solving Notes by In this do the following: We Weight Update then use method to This is very fast solving these based on an 11 of in TCS, 2015” written Eggerling on notes Kaul that we last lecture. last lecture In lecture, we to the order to fairly smartly advice of the game-setting with days and N follows: . expert gives some advice: UP predicts, based on of the expert, or DOWN. with knowledge and aggregator’s the UP/DOWN outcome. observes the outcome suffers his was incorrect. by ε (the i w(1) i to 1. (All experts are equally the ning.) At t: based a weighted majority vote , . w(t) N ). • observing the cost vector, set w(t) i i (Discount Last lecture analyzed the case when ε 1/2. same proof Theorem sequence outcomes, duration and expert i #</code> | <code>Advanced Algorithms March 22, 2022 Lecture 9: Solving LPs using Multiplicative Weights Notes by Ola Svensson1 In this lecture we do the following: • We describe the Multiplicative Weight Update (actually Hedge) method. • We then use this method to solve covering LPs. • This is a very fast and simple (i.e., very attractive) method for solving these LPs approximately. These lecture notes are partly based on an updated version of “Lecture 11 of Topics in TCS, 2015” that were written by Vincent Eggerling and Simon Rodriguez and on the lecture notes by Shiva Kaul that we used in the last lecture. 1 Recall last lecture In the previous lecture, we saw how to use the weighted majority method in order to fairly smartly follow the advice of experts. Recall that the general game-setting with T days and N experts was as follows: For t = 1, . . . , T: 1. Each expert i ∈[N] gives some advice: UP or DOWN 2. Aggregator (you) predicts, based on the advice of the expert, UP or DOWN. 3. Adversary, with k...</code> |
526
+ | <code>analyzed the case when = same the following For any outcomes, T, and of of + O(log(N)/ε) These notes for the have not peer-reviewed and may contain inconsistent omit works.</code> | <code>Last lecture we analyzed the case when ε = 1/2. The same proof gives the following Theorem 1 For any sequence of outcomes, duration T, and expert i ∈[N], # of WM mistakes ≤2(1 + ε) · (# of i’s mistakes) + O(log(N)/ε) . 1Disclaimer: These notes were written as notes for the lecturer. They have not been peer-reviewed and may contain inconsistent notation, typos, and omit citations of relevant works. 1</code> |
527
+ | <code>proof by defining function: each t . Φ(t) = i∈[N] i . We lower the Φ(T +1) using of of then bound terms bound: of goes a mistake initial weight is 1, Φ(T +1) = w(T +1) j ≥w(T +1) = (1 −ε)# of i’s mistakes bound: Every errs, at least half the experts (since weighted weights are (1 follows that down factor (1 WM ≤Φ(1) · WM = (1 WM for the equality that N initialized a weight of 1. The above give us (1 i’s mistakes ≤Φ(T WM . logs on then statement. 2 the game: for randomized strategies In you proved that instances for which weighted twice as as expert! is will to of making prediction (that the adversary then create side note this often is to following days and experts: 1, ,</code> | <code>Proof [Sketch] The proof was done by defining a potential function: for each t = 1, . . . , T + 1, let Φ(t) = X i∈[N] w(t) i . We now lower bound the “final” potential Φ(T +1) using the number of mistakes of i. We then upper bound it in terms of our number of mistakes. Lower bound: The weight of expert i goes down by a factor (1 −ε) for each mistake i does. As the initial weight of i is 1, Φ(T +1) = X j∈[N] w(T +1) j ≥w(T +1) i = (1 −ε)# of i’s mistakes . Upper bound: Every time WM errs, at least half the weight of the experts was wrong (since weighted majority was wrong). These weights are then decreased by (1 −ε). It follows that the potential goes down by at least a factor (1 −ε/2) every time WM errs. And so Φ(T +1) ≤Φ(1) · (1 −ε/2)# of WM mistakes = N · (1 −ε/2)# of WM mistakes , where for the equality we used that Φ(1) = N since each expert was initialized with a weight of 1. The above bounds give us (1 −ε)# of i’s mistakes ≤Φ(T +1) ≤N · (1 −ε/2)# of WM mistakes . Taking logs on b...</code> |
528
+ * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
529
+ ```json
530
+ {
531
+ "scale": 20.0,
532
+ "similarity_fct": "cos_sim"
533
+ }
534
+ ```
535
+
536
+ ### Evaluation Dataset
537
+
538
+ #### Unnamed Dataset
539
+
540
+ * Size: 3,827 evaluation samples
541
+ * Columns: <code>anchor</code> and <code>positive</code>
542
+ * Approximate statistics based on the first 1000 samples:
543
+ | | anchor | positive |
544
+ |:--------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
545
+ | type | string | string |
546
+ | details | <ul><li>min: 15 tokens</li><li>mean: 175.44 tokens</li><li>max: 258 tokens</li></ul> | <ul><li>min: 55 tokens</li><li>mean: 432.79 tokens</li><li>max: 512 tokens</li></ul> |
547
+ * Samples:
548
+ | anchor | positive |
549
+ |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
550
+ | <code>Adrenocorticosteroids & Adrenocortical Antagonists 707 hypertension occurs. or of hydrocortisone growth occurs in and have than steroid at equivalent in larger amounts, such and effects to glucocorticoid effects, sodium and loss potassium. and function, this leads to a hypochloremic alkalosis and eventually to pressure. In patients hypoproteinemia, renal or also In with even degrees to effects be minimized steroids, of potassium supplements. Adrenal administered more may treatment to appropriate at of stress for 24–48 severe to ten-fold dosage increases 48–72 trauma or major corti- costeroid dosage is to it should be slowly. be quite levels. may take 2–12 months hypothalamic-pituitary-adrenal function acceptably, cortisol levels to another months. The suppression is pituitary and treatment ACTH reduce time required the return function. If the in gluco- for certain disorder, the</code> | <code>CHAPTER 39 Adrenocorticosteroids & Adrenocortical Antagonists 707 hypertension also occurs. In dosages of 45 mg/m 2 /d or more of hydrocortisone or its equivalent, growth retardation occurs in children. Medium-, intermediate-, and long-acting glucocorticoids have greater growth-suppressing potency than the natural steroid at equivalent doses. When given in larger than physiologic amounts, steroids such as cortisone and hydrocortisone, which have mineralocorticoid effects in addition to glucocorticoid effects, cause some sodium and fluid retention and loss of potassium. In patients with normal cardiovas- cular and renal function, this leads to a hypokalemic, hypochloremic alkalosis and eventually to a rise in blood pressure. In patients with hypoproteinemia, renal disease, or liver disease, edema may also occur. In patients with heart disease, even small degrees of sodium retention may lead to heart failure. These effects can be minimized by using synthetic non-salt-retaining steroids, ...</code> |
551
+ | <code>is and treatment ACTH not the for of function. If too in corticoids a of the or in intensity. patients an patients Cushing’s also with These symptoms or ing, weight lethargy, joint or pain, postural many of symptoms reflect true deficiency, may occur presence normal even elevated cortisol gesting glucocorticoid Contraindications A. Special glucocorticoids carefully for development of retention peptic osteopo- rosis, should as as and administration alternate-day) should when therapeutic obtained schedule. on relatively low doses of such are intercurrent or acci- dents B. must with great patients with peptic heart heart cer- osteoporosis, or Selection of Dosage Schedule anti- mineralocorticoid duration action, cost, and dosage forms ( these should selecting the drug used. ACTH Adrenocortical Steroids In patients with used</code> | <code>is not a pituitary problem, and treatment with ACTH does not reduce the time required for the return of normal function. If the dosage is reduced too rapidly in patients receiving gluco- corticoids for a certain disorder, the symptoms of the disorder may reappear or increase in intensity. However, patients without an underlying disorder (eg, patients cured surgically of Cushing’s disease) also develop symptoms with rapid reductions in cortico- steroid levels. These symptoms include anorexia, nausea or vomit- ing, weight loss, lethargy, headache, fever, joint or muscle pain, and postural hypotension. Although many of these symptoms may reflect true glucocorticoid deficiency, they may also occur in the presence of normal or even elevated plasma cortisol levels, sug- gesting glucocorticoid dependence. Contraindications & Cautions A. Special Precautions Patients receiving glucocorticoids must be monitored carefully for the development of hyperglycemia, glycosuria, sodium retention with ede...</code> |
552
+ | <code>39–1 these factors be taken into drug be ACTH versus In with normal was used in production of similar effects. However, an the use a has in which was be were probably due amounts of corticosteroids the dosage B. Dosage the to be used, the physician the the of be required desired and of In some required for maintenance of the desired less the initial and the lowest dosage should gradually lowering the a small in is noted. When it is continuously elevated corticosteroid to suppress oral doses required. exists respect use of in inflammatory allergic same in a be effective than given in smaller doses in a Severe autoimmune vital organs must aggressively, and undertreatment is as To deposition immune and leukocytes 1 mg/kg/d predni- required initially. This dosage is main- tained the gradually are required alternate-day the may control in</code> | <code>available ( Table 39–1 ), and these factors should be taken into account in selecting the drug to be used. A. ACTH versus Adrenocortical Steroids In patients with normal adrenals, ACTH was used in the past to induce the endogenous production of cortisol to obtain similar effects. However, except when an increase in androgens is desir- able, the use of ACTH as a therapeutic agent has been abandoned. Instances in which ACTH was claimed to be more effective than glucocorticoids were probably due to the administration of smaller amounts of corticosteroids than were produced by the dosage of ACTH. B. Dosage In determining the dosage regimen to be used, the physician must consider the seriousness of the disease, the amount of drug likely to be required to obtain the desired effect, and the duration of therapy. In some diseases, the amount required for maintenance of the desired therapeutic effect is less than the dose needed to obtain the initial effect, and the lowest possible dosage for th...</code> |
553
+ * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
554
+ ```json
555
+ {
556
+ "scale": 20.0,
557
+ "similarity_fct": "cos_sim"
558
+ }
559
+ ```
560
+
561
+ ### Training Hyperparameters
562
+ #### Non-Default Hyperparameters
563
+
564
+ - `eval_strategy`: steps
565
+ - `per_device_train_batch_size`: 128
566
+ - `per_device_eval_batch_size`: 128
567
+ - `learning_rate`: 2e-05
568
+ - `num_train_epochs`: 5
569
+ - `warmup_ratio`: 0.1
570
+ - `fp16`: True
571
+ - `dataloader_drop_last`: True
572
+ - `dataloader_num_workers`: 2
573
+ - `load_best_model_at_end`: True
574
+ - `push_to_hub`: True
575
+ - `hub_model_id`: RikoteMaster/embedder-granite
576
+ - `hub_strategy`: end
577
+ - `hub_private_repo`: True
578
+
579
+ #### All Hyperparameters
580
+ <details><summary>Click to expand</summary>
581
+
582
+ - `overwrite_output_dir`: False
583
+ - `do_predict`: False
584
+ - `eval_strategy`: steps
585
+ - `prediction_loss_only`: True
586
+ - `per_device_train_batch_size`: 128
587
+ - `per_device_eval_batch_size`: 128
588
+ - `per_gpu_train_batch_size`: None
589
+ - `per_gpu_eval_batch_size`: None
590
+ - `gradient_accumulation_steps`: 1
591
+ - `eval_accumulation_steps`: None
592
+ - `torch_empty_cache_steps`: None
593
+ - `learning_rate`: 2e-05
594
+ - `weight_decay`: 0.0
595
+ - `adam_beta1`: 0.9
596
+ - `adam_beta2`: 0.999
597
+ - `adam_epsilon`: 1e-08
598
+ - `max_grad_norm`: 1.0
599
+ - `num_train_epochs`: 5
600
+ - `max_steps`: -1
601
+ - `lr_scheduler_type`: linear
602
+ - `lr_scheduler_kwargs`: {}
603
+ - `warmup_ratio`: 0.1
604
+ - `warmup_steps`: 0
605
+ - `log_level`: passive
606
+ - `log_level_replica`: warning
607
+ - `log_on_each_node`: True
608
+ - `logging_nan_inf_filter`: True
609
+ - `save_safetensors`: True
610
+ - `save_on_each_node`: False
611
+ - `save_only_model`: False
612
+ - `restore_callback_states_from_checkpoint`: False
613
+ - `no_cuda`: False
614
+ - `use_cpu`: False
615
+ - `use_mps_device`: False
616
+ - `seed`: 42
617
+ - `data_seed`: None
618
+ - `jit_mode_eval`: False
619
+ - `use_ipex`: False
620
+ - `bf16`: False
621
+ - `fp16`: True
622
+ - `fp16_opt_level`: O1
623
+ - `half_precision_backend`: auto
624
+ - `bf16_full_eval`: False
625
+ - `fp16_full_eval`: False
626
+ - `tf32`: None
627
+ - `local_rank`: 0
628
+ - `ddp_backend`: None
629
+ - `tpu_num_cores`: None
630
+ - `tpu_metrics_debug`: False
631
+ - `debug`: []
632
+ - `dataloader_drop_last`: True
633
+ - `dataloader_num_workers`: 2
634
+ - `dataloader_prefetch_factor`: None
635
+ - `past_index`: -1
636
+ - `disable_tqdm`: False
637
+ - `remove_unused_columns`: True
638
+ - `label_names`: None
639
+ - `load_best_model_at_end`: True
640
+ - `ignore_data_skip`: False
641
+ - `fsdp`: []
642
+ - `fsdp_min_num_params`: 0
643
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
644
+ - `fsdp_transformer_layer_cls_to_wrap`: None
645
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
646
+ - `deepspeed`: None
647
+ - `label_smoothing_factor`: 0.0
648
+ - `optim`: adamw_torch
649
+ - `optim_args`: None
650
+ - `adafactor`: False
651
+ - `group_by_length`: False
652
+ - `length_column_name`: length
653
+ - `ddp_find_unused_parameters`: None
654
+ - `ddp_bucket_cap_mb`: None
655
+ - `ddp_broadcast_buffers`: False
656
+ - `dataloader_pin_memory`: True
657
+ - `dataloader_persistent_workers`: False
658
+ - `skip_memory_metrics`: True
659
+ - `use_legacy_prediction_loop`: False
660
+ - `push_to_hub`: True
661
+ - `resume_from_checkpoint`: None
662
+ - `hub_model_id`: RikoteMaster/embedder-granite
663
+ - `hub_strategy`: end
664
+ - `hub_private_repo`: True
665
+ - `hub_always_push`: False
666
+ - `gradient_checkpointing`: False
667
+ - `gradient_checkpointing_kwargs`: None
668
+ - `include_inputs_for_metrics`: False
669
+ - `include_for_metrics`: []
670
+ - `eval_do_concat_batches`: True
671
+ - `fp16_backend`: auto
672
+ - `push_to_hub_model_id`: None
673
+ - `push_to_hub_organization`: None
674
+ - `mp_parameters`:
675
+ - `auto_find_batch_size`: False
676
+ - `full_determinism`: False
677
+ - `torchdynamo`: None
678
+ - `ray_scope`: last
679
+ - `ddp_timeout`: 1800
680
+ - `torch_compile`: False
681
+ - `torch_compile_backend`: None
682
+ - `torch_compile_mode`: None
683
+ - `include_tokens_per_second`: False
684
+ - `include_num_input_tokens_seen`: False
685
+ - `neftune_noise_alpha`: None
686
+ - `optim_target_modules`: None
687
+ - `batch_eval_metrics`: False
688
+ - `eval_on_start`: False
689
+ - `use_liger_kernel`: False
690
+ - `eval_use_gather_object`: False
691
+ - `average_tokens_across_devices`: False
692
+ - `prompts`: None
693
+ - `batch_sampler`: batch_sampler
694
+ - `multi_dataset_batch_sampler`: proportional
695
+
696
+ </details>
697
+
698
+ ### Training Logs
699
+ | Epoch | Step | Training Loss | Validation Loss |
700
+ |:---------:|:--------:|:-------------:|:---------------:|
701
+ | 0.1859 | 50 | 0.3888 | - |
702
+ | 0.3717 | 100 | 0.1835 | - |
703
+ | 0.5576 | 150 | 0.0817 | - |
704
+ | 0.7435 | 200 | 0.0401 | 0.0351 |
705
+ | 0.9294 | 250 | 0.0376 | - |
706
+ | 1.1152 | 300 | 0.0332 | - |
707
+ | 1.3011 | 350 | 0.028 | - |
708
+ | 1.4870 | 400 | 0.0285 | 0.0162 |
709
+ | 1.6729 | 450 | 0.0246 | - |
710
+ | 1.8587 | 500 | 0.0239 | - |
711
+ | 2.0446 | 550 | 0.0241 | - |
712
+ | 2.2305 | 600 | 0.0237 | 0.0130 |
713
+ | 2.4164 | 650 | 0.0222 | - |
714
+ | 2.6022 | 700 | 0.019 | - |
715
+ | 2.7881 | 750 | 0.0235 | - |
716
+ | 2.9740 | 800 | 0.0266 | 0.0120 |
717
+ | 3.1599 | 850 | 0.0214 | - |
718
+ | 3.3457 | 900 | 0.024 | - |
719
+ | 3.5316 | 950 | 0.0249 | - |
720
+ | 3.7175 | 1000 | 0.0213 | 0.0113 |
721
+ | 3.9033 | 1050 | 0.0233 | - |
722
+ | 4.0892 | 1100 | 0.0213 | - |
723
+ | 4.2751 | 1150 | 0.0202 | - |
724
+ | **4.461** | **1200** | **0.0227** | **0.0109** |
725
+ | 4.6468 | 1250 | 0.0229 | - |
726
+ | 4.8327 | 1300 | 0.0196 | - |
727
+
728
+ * The bold row denotes the saved checkpoint.
729
+
730
+ ### Framework Versions
731
+ - Python: 3.10.17
732
+ - Sentence Transformers: 4.1.0
733
+ - Transformers: 4.52.3
734
+ - PyTorch: 2.7.0+cu126
735
+ - Accelerate: 1.7.0
736
+ - Datasets: 3.6.0
737
+ - Tokenizers: 0.21.1
738
+
739
+ ## Citation
740
+
741
+ ### BibTeX
742
+
743
+ #### Sentence Transformers
744
+ ```bibtex
745
+ @inproceedings{reimers-2019-sentence-bert,
746
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
747
+ author = "Reimers, Nils and Gurevych, Iryna",
748
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
749
+ month = "11",
750
+ year = "2019",
751
+ publisher = "Association for Computational Linguistics",
752
+ url = "https://arxiv.org/abs/1908.10084",
753
+ }
754
+ ```
755
+
756
+ #### MultipleNegativesRankingLoss
757
+ ```bibtex
758
+ @misc{henderson2017efficient,
759
+ title={Efficient Natural Language Response Suggestion for Smart Reply},
760
+ author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
761
+ year={2017},
762
+ eprint={1705.00652},
763
+ archivePrefix={arXiv},
764
+ primaryClass={cs.CL}
765
+ }
766
+ ```
767
+
768
+ <!--
769
+ ## Glossary
770
+
771
+ *Clearly define terms in order to be accessible across audiences.*
772
+ -->
773
+
774
+ <!--
775
+ ## Model Card Authors
776
+
777
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
778
+ -->
779
+
780
+ <!--
781
+ ## Model Card Contact
782
+
783
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
784
+ -->
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "XLMRobertaModel"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": null,
8
+ "eos_token_id": 2,
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": 514,
16
+ "model_type": "xlm-roberta",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 6,
19
+ "pad_token_id": 1,
20
+ "position_embedding_type": "absolute",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.52.3",
23
+ "type_vocab_size": 2,
24
+ "use_cache": true,
25
+ "vocab_size": 250002
26
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "4.1.0",
4
+ "transformers": "4.52.3",
5
+ "pytorch": "2.7.0+cu126"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": "cosine"
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d63dc9c56fb09e54dc9a37bc41e80aad34770dc63a83a803d2e067c0ad9a7b40
3
+ size 427988744
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
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a8d0b7573869188be52cca17a27a84f3cfbc0a5536c28ee1eca82903e8c68c6
3
+ size 17083051
tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "extra_special_tokens": {},
49
+ "mask_token": "<mask>",
50
+ "model_max_length": 512,
51
+ "pad_token": "<pad>",
52
+ "sep_token": "</s>",
53
+ "sp_model_kwargs": {},
54
+ "tokenizer_class": "XLMRobertaTokenizer",
55
+ "unk_token": "<unk>"
56
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:697607fc6ba410ea01355bf4a4f73d1fd301a942e36a7d759a3484ac0ef46a85
3
+ size 6161
training_metadata.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model": "sentence-transformers/all-MiniLM-L6-v2",
3
+ "training_corpus_size": 63474,
4
+ "training_examples": 63474,
5
+ "train_examples": 57126,
6
+ "validation_examples": 6348,
7
+ "validation_split": 0.1,
8
+ "chunk_size": 256,
9
+ "chunk_overlap": 50,
10
+ "batch_size": 128,
11
+ "epochs": 5,
12
+ "learning_rate": 2e-05,
13
+ "warmup_ratio": 0.1,
14
+ "early_stopping_patience": 5,
15
+ "eval_steps": 200,
16
+ "device": "cuda",
17
+ "training_api": "SentenceTransformers_v3+",
18
+ "loss_function": "MultipleNegativesRankingLoss",
19
+ "eval_strategy": "steps",
20
+ "fp16": true,
21
+ "save_strategy": "steps",
22
+ "save_steps": 200,
23
+ "logging_steps": 50,
24
+ "save_total_limit": 3,
25
+ "push_to_hub": true,
26
+ "hub_model_id": "RikoteMaster/retriever_pdf_and_books",
27
+ "hub_private": false
28
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff