metadata
language:
- en
license: apache-2.0
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- dense
- generated_from_trainer
- dataset_size:1179
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
base_model: nlpaueb/legal-bert-base-uncased
widget:
- source_sentence: >-
Mercer International Inc. v. Government of Canada Canada’s
Counter-Memorial August 22, 2014 concerns that such transaction could
result in an increase of its costs of service should Howe Sound purchase
additional electricity so that it could sell its existing self- generation
into the market.266 115. On February 23, 2001, BC Hydro wrote to the BCUC
advising it that some of its customers with self-generation capability
wished to sell power they generate at market prices. BC Hydro requested
that the BCUC initiate a process beginning with a workshop to determine
the extent to which BC Hydro would remain obligated to serve industrial
customers who wished to take their self-generation output to the
market.267 116. Howe Sound, which had significantly decreased its
generation in response to peaking natural gas prices,268 proposed “to
utilize only that part of its generation capacity which [was] idle” and
that “[a]ll of the generation utilized for market sales [would] be
incremental and [would] not require BC Hydro to deliver any additional
electricity to Howe Sound.”269 266 Letter from Craig Folkestad to Jerry
Peet, Re: Howe Sound Pulp and Paper (HSPP) Power Export Opportunities, 12
February 2001 at 1, R-79. (“However, I would be less than candid if I did
not tell you that the management of BC Hydro does, and most likely the
government as its shareholder, will have serious concerns about any
proposal that will see customer self-generated power sold into the market,
and with BC Hydro then being required to supply make-up power under
Schedule 1821. This will be financially detrimental to BC Hydro and its
other ratepayers, both in the short and long term.”); BC Hydro, Letter to
the BCUC, in the Matter of British Columbia Hydro and Power Authority
Obligation to Serve Rate Schedule 1821 Customers with Self-Generation
Capability, 23 February 2001 (“BC Hydro’s 23 February 2001 Letter to the
BCUC”), R-81. See also Pierre Lamarche Statement, ¶¶ 28, 30 (“Howe Sound
agreed with BC Hydro that such arbitrage could have a negative effect on
BC Hydro ratepayers, but that self- generators should have the ability to
sell incremental or idle self-generation”); Lester Dyck Statement, ¶ 36;
Jim Scouras Statement, ¶ 21. 267 BC Hydro’s 23 February 2001 Letter to the
BCUC, R-81. 268 See Pierre Lamarche Statement, ¶¶ 23-26. Howe Sound was,
in fact, considering shutting down its condensing turbine completely: ¶
26. 269 Howe Sound Pulp and Paper, Letter to the BCUC, in the Matter of
British Columbia Hydro and Power Authority Obligation to Serve Rate
Schedule 1821 Customers with Self-Generation Capability, 27 February 2001
at bates 144039-144040, R-80. 63
sentences:
- >-
Mr. Switlishoff, can you clarify the implications of the EPA termination
with BC Hydro in 2020? Did the Claimant indeed assume that electricity
would be purchased beyond that term?
Certainly. The Claimant did not assume purchase beyond 2020. They fully
anticipated market conditions shifting and prepared for alternate sales
strategies after the EPA's conclusion.
Interesting, because earlier records indicated that one-third of their
damages calculation was based on perpetual purchases by BC Hydro. Are
you saying that this wasn't part of their original strategy?
Yes, that's correct. The calculations were hypothetical and never
incorporated into any actionable business strategy by the Claimant.
Regarding the BCUC orders, is it accurate to state that Order G-48-09
imposed restrictions that led to financial losses for the Claimant?
No, Order G-48-09 didn’t cause any financial damage as the Claimant had
alternative arrangements for selling their electricity, including
tapping into different markets and agreements.
- >-
Mr. Friesen, can you confirm whether Celgar had any realistic
opportunities to sell its self-generated electricity to regions outside
of British Columbia in 2008?
Yes, we had identified several potential buyers during that time who
were interested in our output at competitive rates.
Isn't it true that Celgar struggled to secure transmission access for
exporting this electricity, making such sales challenging?
Actually, we had preliminary agreements lined up which would have
ensured us the necessary transmission access to conduct these sales
effectively.
But wouldn’t any potential sales have been economically inefficient,
given the high cost of replacement electricity from FortisBC?
Our analysis showed that the revenues from these sales would indeed
cover the costs and provide a margin, contrary to what was suggested.
- >-
Ms. Peet, can you explain how the proposal was initiated and who was
involved in the discussions with BC Hydro?
Certainly. I worked alongside representatives from our Technical
Department and coordinated with Mr. Jerry Peet, who led the discussions.
Mr. Peet, together with Craig Folkestad, our Key Account Manager,
presented the compiled data to BC Hydro, where they discussed the
proposals extensively before reaching an agreement on the thresholds.
Thank you. Now, regarding Howe Sound's generation strategy during the
period of gas price peaks, how did you propose to manage your generation
capacity?
During this time, Howe Sound proposed to utilize only the part of our
generation capacity that was idle. Any generation incrementally used for
market sales would not necessitate additional electricity delivery from
BC Hydro to our facilities.
Could you clarify if BC Hydro had concerns about this proposal affecting
costs and obligations?
Yes, BC Hydro did express concerns that selling self-generated power
into the market might increase its service costs. They worried it could
impact obligations under Rate Schedule 1821 and potentially negatively
affect other ratepayers.
- source_sentence: >-
619. Moreover, the restriction on below-GBL sales to third parties was not
otherwise necessary to BC Hydro’s Bioenergy Phase I procurement, as
demonstrated both by the fact that BC Hydro had at least, at one point in
the EPA negotiations with Celgar, agreed not to include the
restriction,707 and the fact that the BCUC set a GBL for Tolko in 2001
that restricted below-GBL sales completely outside the context of any
procurement.708 As the Tribunal will recall, the GBL concept originated in
BCUC Order G-38-01 to address Howe Sound’s desire to sell power to
California. It has no necessary relationship to any BC or BC Hydro
procurement. 620. Mercer agrees with Canada and the ADF tribunal that
“procurement” refers to “the obtaining by purchase by a governmental
agency or entity of title to . . . possession of, for instance, goods,
supplies, materials and machinery.”709 But BC Hydro did not obtain any
good or service through the challenged restriction on sales to
third-parties. At issue is Celgar’s below- load self-generated electricity
that BC Hydro declined to buy. The measures restricted Celgar from
providing, to anyone. Under Canada’s preferred definition, that is not
procurement. 621. The ADF case does not suggest otherwise. In ADF, a
cabinet-level agency of the Commonwealth of Virginia (the Department of
Transportation) was responsible for “the construction of a multi-phased
project designed to improve the safety and efficiency of” a major highway
system in the area of Springfield, Virginia, near Washington, DC.710 The
project included the construction of ramps and bridges curving above the
relevant highways, as well as of 707 See supra ¶ 38 and n.28. 708 See
Memorial, ¶¶ 240–47. 709 CA-1, ADF (NAFTA), ¶ 161; Counter-Memorial, ¶
342. See also CA-16, UPS II (NAFTA), ¶ 135 (concluding that a Postal
Imports Agreement in which the Canadian customs authority obtains
materials handling, data entry, and duty collection services, is a
procurement). 710 CA-1, ADF (NAFTA), ¶ 45. 304
sentences:
- >-
Can you explain the Tribunal's final stance concerning the Claimant's
claims under the 2009 EPA and NAFTA Articles?
Certainly. The Tribunal, by a majority, decided it had no jurisdiction
over the Claimant's claims under NAFTA Articles 1102, 1103, and 1105,
except for those alleging discriminatory treatment under Article 1105.
So, claims for compensation and related interest were dismissed.
And how did the Tribunal address the Claimant's request for a
Supplementary Decision under the ICSID Additional Facility Rules?
The Claimant requested a Supplementary Decision regarding alleged
discrimination under NAFTA Articles 1102 and 1103 related to BCUC Order
G-48-09. However, the Respondent denied this request, and the Tribunal's
handling was complicated by the passing of Professor Orrego Vicuña.
Was there any consensus among the Tribunal members on handling the
Claimant's request before Professor Orrego Vicuña's passing?
Yes, there was. All three Tribunal members reached a consensus during a
conference call. This was before Professor Orrego Vicuña became unable
to sign the final document.
- >-
Could you clarify whether Celgar's Energy Project Certificate had any
ongoing effects after the legislative changes in 1995?
Yes, Celgar's Energy Project Certificate continued to be recognized, but
it wasn't explicitly covered under the updated Environmental Assessment
Act after 1995. The transitional provisions didn't apply clearly to
pre-existing orders.
But isn't it true that prior orders like Celgar's were explicitly deemed
to have continued under the new Act?
No, the Ministers' Orders required separate re-evaluation before being
reaffirmed under the new legislation. This wasn’t automatic for older
orders.
Regarding FortisBC's access principles, did they apply to
self-generators like Celgar?
Initially, self-generators weren’t considered under those principles. It
took several years before any provisions applied to them.
- >-
Mr. Smith, during the negotiations for the EPA, can you confirm whether
BC Hydro pushed for a longer contract term with Celgar?
Yes, BC Hydro did suggest a longer contract term, but we never received
a formal request for anything more than 15 years.
Interesting, because it has been indicated that BC Hydro was seeking at
least a 20-year term. Are you certain about your statement?
I understand that’s what it might seem from other discussions, but the
formal conversations we had revolved around 15 years as the maximum
offered.
Regarding the restrictions on selling generated electricity, did these
originate from BC Hydro’s procurement process?
Actually, the restrictions coincided with initial procurement
discussions, suggesting they were integral to the process.
- source_sentence: >-
its witness, Mr. Dyck, confirmed that information regarding BC Hydro’s
treatment of other pulp mills was never shared with Mercer.59 Mercer only
acquired constructive knowledge of its comparators’ treatment through its
counsel during the document production phase of these proceedings in May
2013. 34. As established above, moreover, Mercer could not have acquired
knowledge of loss or damage, at the earliest, until the GBL-based
exclusivity provisions were either final or in effect. As noted, the
exclusivity provision at issue did not take effect, under the terms of the
EPA, until the Commercial Operation date of 27 September 2010, and it did
not become final until the BCUC ruled in Order G-48- 09 against Celgar’s
attempt to purchase electricity from FortisBC while selling power. Both of
these dates are within the period of limitations; thus, Mercer’s Minimum
Standard of Treatment claim is within the period of limitations.60 II. THE
MINISTERS’ ORDER IMPOSES NO SELF-SUPPLY OBLIGATION OR ELECTRICITY SALES
RESTRICTION ON CELGAR 35. During the hearing, Canada all but abandoned its
Ministers’ Order argument. Canada’s relative silence on this issues was
understandable, because (i) the parties’ legal experts agree that the
language in the Ministers’ Order must be clear and unambiguous in order to
impose a binding legal obligation on Celgar that restricts its right to
sell electricity,61 and (ii) Canada’s witnesses confirmed that there
simply is no clear and consistent language in the 1991 Ministers’ Order
that imposed any self-supply or load displacement obligation, or otherwise
restricted Celgar’s right to sell its self-generated 59 See supra, Section
I.C.1. 60 See supra, Section I.C.1. 61 See Expert Report of David Austin
(14 December 2014) (“Austin Expert Report”) ¶¶ 21-30; Expert Report of
David Bursey (28 March 2015) (“Bursey Expert Report”) ¶¶ 182-186, 191 (Mr.
Bursey asserts that the language of the Ministers’ Order is clear; he does
not refute the general principle that the language of the Ministers’ Order
would need to be clear and unambiguous to restrict Celgar’s right to sell
electricity); Mercer Letter to Tribunal pp. 9-10 (12 July 2015); Reply ¶¶
57, 94-101. - 17 -
sentences:
- >-
Mr. Smith, can you clarify how BC Hydro determined the GBL for Celgar
compared to other mills?
Certainly. BC Hydro based Celgar’s GBL on one year of operational data
from 2007, even though they led us to believe they would consider an
average of three years. Other mills were not subject to the same method,
which BC Hydro failed to communicate to us.
Was any consideration given to the economic or financial performance of
Celgar during the GBL determination?
No, BC Hydro never indicated that such data would be relevant. They
didn't request any economic or financial information about Celgar’s
operations at any time during the process.
How does this approach contrast with the treatment of other mills, like
Skookumchuck?
The Skookumchuck mill operated as an independent power producer and had
more flexibility with their agreements. Celgar, however, was integrated
into its recovery boiler operations and was treated less favorably
despite assurances from BC Hydro.
- >-
Mr. Scouras, can you clarify how the California Energy Crisis impacted
BC Hydro's power acquisition strategy?
Certainly. The California Energy Crisis significantly influenced the
strategy. Following the crisis, BC Hydro was compelled to secure a more
reliable power supply, which led to initiatives like the 2002
Customer-Based Generation Call for Power, as outlined in BCUC Order
G-38-01 and the 2002 Energy Plan.
And how did these efforts evolve by the time of the 2007 Energy Plan?
The 2007 Energy Plan introduced the Bioenergy Strategy and the Bioenergy
Call for Power – Phase I. This was part of a move towards sustainable
energy sources, leveraging biomass projects such as the Celgar Mill's
Biomass Realization Project.
Speaking of Celgar, what was the structure of their agreement with BC
Hydro?
Celgar entered a 2009 Energy Purchase Agreement with BC Hydro,
complemented by a Side Letter Agreement. This arrangement included
specific provisions for seller-consumed eligible electricity, a key
component in their integration with BC Hydro’s power acquisition
framework.
- >-
Is it true that Celgar has restrictions when it comes to selling its
self-generated electricity below its load?
Yes, that's correct. According to BCUC Order G-48-09, Celgar is
prevented from obtaining energy from FortisBC while selling
self-generated electricity below its load.
Can you clarify how the agreements with BC Hydro affect Celgar's ability
to sell electricity?
The GBL-related provisions in Celgar’s 2009 EPA with BC Hydro preclude
the mill from selling energy below its 2007 load to any third party.
Essentially, this strands Celgar’s below-GBL self-generated electricity,
requiring them to self-supply the first 349 GWh/year of its own load.
Was there any legal obligation imposed by the 1991 Ministerial Order
that affected this arrangement?
No, there was no ongoing legal obligation from the 1991 Ministerial
Order for Celgar to self-supply or restrict its electricity sales based
on that order.
- source_sentence: >-
7.74 Mr Merwin (of Celgar) proclaimed Order G-188-1 to be a “major
victory” at the time in his memorandum of 7 December 2011 to Mercer’s
Board of Directors.288 He stated that the BCUC had confirmed that “Celgar
is able to buy all of its power requirements from FortisBC and free to
sell the output of all of its generation to third parties.”289 7.75 This
interpretation of Order G-188-1 was confirmed by the BCUC in its
subsequent Decision of 27 December 2012 accompanying Order G-202-12. It
summarised the entitlements of customers of FortisBC: “[The] entitlement
to non-BC Hydro PPA embedded cost power by a self-generating customer may
be as high as 100 percent of load as nominated by that customer”.290 (H)
The Tribunal’s Analysis on BCUC Order G-48-09 7.76 In the Tribunal’s view,
on the evidence before it, the Claimant falls short of establishing that
BCUC Order G-48-09 or any associated aspect of the BCUC’s regulatory
regime breaches the customary international law standard of treatment
under NAFTA Article 1105(1), as explained in the NAFTA award in Merill &
Ring v Canada. The Claimant has not established irrationality, injustice,
arbitrariness, or a violation of due process within the meaning of the
customary international law standard. 7.77 As to transparency, it suffices
to cite the Cargill Award cited above, in which the tribunal decided that
the customary international law standard had not yet been shown to embrace
a claim to transparency.291 The Tribunal also notes that the tribunal in
Merill & Ring decided that transparency was not part of the customary
international law standard.292 In any event, even if applicable, the
Tribunal would not be inclined to decide that the Claimant’s case reaches
the threshold for non-transparency. 288 Memorandum from Management to
Mercer International Board of Directors, Re Update on Celgar’s Generator
Baseline Issue of 7 December 2011, p. 1 [R-531] (emphasis in the
original). 289 Id. 290 BCUC Decision and Order No. G-202-12 of December
27, 2012 [R-265], p. 3. 291 Cargill v. Mexico, ibid, Paragraphs 290 and
294. 292 Merill & Ring v Canada, ibid, Paragraph 208.
sentences:
- >-
Mr. Merwin, could you explain the impact of Order G-188-1 on Celgar's
operations?
Certainly. Order G-188-1 was indeed a significant development for
Celgar. It allowed us to purchase all our power needs from FortisBC
while being free to sell the entirety of our generated power to third
parties. This was confirmed by the BCUC's decision later in 2012.
And how did these regulatory changes align with your steam savings and
energy projects?
Well, at that time, we were already pursuing projects to improve steam
utilization and energy production. We identified multiple PINCH projects
to enhance efficiency and planned a retrofit of our power boiler to
generate more steam. The changes allowed us to leverage surplus steam
for our Green Energy Project, aiming to install a 48 MW condensing
turbine.
Did these projects have any influence on the discussions with BC Hydro
regarding Celgar’s GBL settings?
Yes, they did. There were some concerns from our side regarding the
years used to set Celgar's GBL. We preferred that BC Hydro considered
the years post-2005, reflecting higher pulp production and the
efficiencies we had achieved through projects like Blue Goose, which
influenced our 2007 operations.
- >-
Mr. Jones, could you clarify the nature of the Ministry's involvement
with the rules governing self-generation for FortisBC’s service area?
Certainly. The Ministry decided to monitor the BCUC proceedings on
FortisBC's compliance filing but did not take an active role. They did
not intervene in these proceedings.
Are you saying there was no intervention despite concerns about
self-generation rules?
That's correct. Although there were discussions, the Ministry's
involvement did not go beyond consultations and offering informal
feedback.
And regarding Celgar's dealings with BC Hydro, were the GBL
methodologies uniformly applied?
While the process was meant to be consistent, BC Hydro's methodology
varied slightly for Celgar due to unique circumstances not present with
other mills like Tolko.
So, you're saying Tolko's situation was not comparable?
Yes, each mill’s situations were distinct due to operational
differences, and Celgar's treatment was unique to its operational needs,
which were unlike Tolko's.
- >-
Mr. Doe, can you explain the purpose of the GBL assigned to Celgar in
the EPA with BC Hydro?
Certainly. The GBL was set to ensure Celgar could not sell electricity
to third parties at prices below market rates, which aligns with BC
Hydro's procurement strategy to secure low-cost power.
Isn't it true that this restriction on selling to third parties was more
about preventing arbitrage than aligning procurement?
No, the main goal was to control market prices through BC Hydro's
procurement process, preventing excess low-cost electricity from
flooding the market.
But wasn't this restriction actually imposed to protect ratepayers due
to concerns over increased embedded cost power from BC Hydro customers
like Celgar?
The primary focus was always to prevent market destabilization, rather
than just ratepayer protection. BC Hydro wanted to manage their supply
efficiently.
- source_sentence: >-
Clause (ii) explicitly required BC Hydro to treat as incremental and
eligible for procurement “existing” generation from already “installed
capacity” that “has been sold to third parties.” When asked why
electricity Celgar had been selling to Northpoint and FortisBC under
existing and terminable contracts did not qualify as “incremental
generation” under the very terms of Addendum 8, Mr. Dyck responded that
Addendum 8 “is not my document. This is Power Acquisition’s document.”17
Mr. Dyck thus understood that his task encompassed more than just power
acquisition. He then stated that, for Celgar, he followed his own
“interpretation,” one of “determining what was incremental to what had
been generated.”18 This interpretation, of course, flatly is inconsistent
with Addendum 8, which specifically defined “what had been generated” as
eligible, incremental power as long as it had been sold to third-parties
and not used for self-supply. Canada cannot claim that Celgar’s GBL-based
sales prohibition is purely procurement-related when it departs from BC
Hydro’s own procurement specifications. 11. Too, Canada’s contention that
the prohibition on below-GBL sales to third-parties is procurement-related
because it is necessary to assure BC Hydro “security of supply” is
fatuous. BC Hydro’s Mr. Scouras claimed that, without the provision, a
proponent could elect to sell electricity promised to BC Hydro to a
third-party instead.19 But Celgar’s promise to supply 238 GWh/yr of firm
electricity to BC Hydro already effectively precludes it from selling that
electricity to a third-party, as 16 R-121, BC Hydro Bioenergy Call for
Power (Phase 10 Addendum 8 (7 May 2008), p. 4, § 8 (emphasis added). See
also Scouras First Witness Statement, ¶ 44 (explaining that the “Existing
Contract” language meant that the existing contract could lawfully be
terminated prior to the Commercial Operation Date in the EPA.). 17 L.
Dyck, Tr. 1487:13-14. 18 L. Dyck, Tr. 1490:3-4. 19 Scouras Second Witness
Statement, ¶ 8; Rejoinder, ¶ 215. - 6 -
sentences:
- >-
Mr. Stockard, can you confirm the baseline year used by BC Hydro for
Celgar’s generation baseline?
Yes, the baseline year used was 2007, which BC Hydro established to
address procurement policies and incentivize new generation.
But isn't it true that a 2006 baseline would have been more consistent
with previous orders, like Order G-38-01?
I don't believe so. The use of a 2007 baseline accurately reflected the
conditions in line with BC policies at that time.
Isn’t Celgar restricted from selling below its generation baseline, even
to third parties?
Actually, Celgar isn’t imposed with such restrictions under the EPA. The
terms are more aligned with BC Hydro’s procurement scope.
According to documentation, the GBL expressly limits below-GBL sales to
third parties, doesn’t it?
That's not my understanding. The GBL provisions were purely for aligning
purchase commitments with Celgar’s production capabilities.
- >-
Mr. Merwin, can you clarify your understanding of the term 'normal
operations' as it pertains to the agreements you had with BC Hydro?
Certainly. At the time, I understood 'normal operations' to mean what
our usual electricity production levels were, with some flexibility for
unforeseen changes. We believed this would be adjusted in our agreements
accordingly.
According to Mr. Dyck, there was no confusion on your end regarding
'normal operations', yet you are claiming otherwise. Can you explain
this discrepancy?
I recall there was definitely some confusion on our side. We asked for
further clarification on several occasions, but the responses were
vague. It's possible Mr. Dyck might not remember all his conversations
accurately.
And when it comes to the GBL set during the 2009 EPA, would you say BC
Hydro overstepped by imposing a self-supply obligation on Celgar?
Not exactly. The self-supply obligation was something we expected as
part of our arrangement with BC Hydro. It was standard procedure, and we
were fully prepared to adhere to it.
- >-
Mr. Dyck, during the negotiations for Celgar's agreement with BC Hydro,
was there any discussion about selling power to third parties before the
agreement was finalized?
Yes, there were discussions about the possibility, but the agreement
ultimately allowed Celgar to sell all its existing capacity to third
parties.
Are you saying the agreement did not restrict below-GBL sales to third
parties?
That's correct. The final agreement did not impose any such
restrictions. It focused primarily on ensuring Celgar's supply
commitments to BC Hydro.
And what about the changes made in November 2008 regarding those sales
provisions? Are you aware of any alterations affecting third-party
agreements?
To my knowledge, the November 2008 adjustments did not impact our
ability to sell to third parties under the GBL.
Just to clarify, are you stating that there was no modification that
introduced a restriction on below-GBL sales?
Correct, there was no such modification in the agreement.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
model-index:
- name: RAG legal-BERT CEAT
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 768
type: dim_768
metrics:
- type: cosine_accuracy@1
value: 0.09090909090909091
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.22727272727272727
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.2727272727272727
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.3409090909090909
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.09090909090909091
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.07575757575757576
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.05454545454545456
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.03409090909090909
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.09090909090909091
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.22727272727272727
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.2727272727272727
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.3409090909090909
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.21022357016371263
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.1689183501683502
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.17951884296669934
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 512
type: dim_512
metrics:
- type: cosine_accuracy@1
value: 0.07575757575757576
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.2196969696969697
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.25757575757575757
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.3409090909090909
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.07575757575757576
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.07323232323232322
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.05151515151515152
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.03409090909090909
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.07575757575757576
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.2196969696969697
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.25757575757575757
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.3409090909090909
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.1991932843140744
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.15463864838864838
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.16490658649101975
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 256
type: dim_256
metrics:
- type: cosine_accuracy@1
value: 0.08333333333333333
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.21212121212121213
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.25757575757575757
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.3409090909090909
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.08333333333333333
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.0707070707070707
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.05151515151515152
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.03409090909090909
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.08333333333333333
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.21212121212121213
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.25757575757575757
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.3409090909090909
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.20170243937575938
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.15850168350168348
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.1701274080868296
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 128
type: dim_128
metrics:
- type: cosine_accuracy@1
value: 0.08333333333333333
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.18181818181818182
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.24242424242424243
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.3106060606060606
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.08333333333333333
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.0606060606060606
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.048484848484848485
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.03106060606060606
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.08333333333333333
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.18181818181818182
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.24242424242424243
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.3106060606060606
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.18512158083530794
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.14629629629629629
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.15809981777726995
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: dim 64
type: dim_64
metrics:
- type: cosine_accuracy@1
value: 0.05303030303030303
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.13636363636363635
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.18181818181818182
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.2727272727272727
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.05303030303030303
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.045454545454545456
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.03636363636363637
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.02727272727272728
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.05303030303030303
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.13636363636363635
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.18181818181818182
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.2727272727272727
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.1503390669056788
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.11273749398749398
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.12617234669895142
name: Cosine Map@100
RAG legal-BERT CEAT
This is a sentence-transformers model finetuned from nlpaueb/legal-bert-base-uncased on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
Model Details
Model Description
- Model Type: Sentence Transformer
- Base model: nlpaueb/legal-bert-base-uncased
- Maximum Sequence Length: 512 tokens
- Output Dimensionality: 768 dimensions
- Similarity Function: Cosine Similarity
- Training Dataset:
- json
- Language: en
- License: apache-2.0
Model Sources
- Documentation: Sentence Transformers Documentation
- Repository: Sentence Transformers on GitHub
- Hugging Face: Sentence Transformers on Hugging Face
Full Model Architecture
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
(1): Pooling({'word_embedding_dimension': 768, '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})
)
Usage
Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
pip install -U sentence-transformers
Then you can load this model and run inference.
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sirtobsi/ceat-fc-rag")
# Run inference
sentences = [
'Clause (ii) explicitly required BC Hydro to treat as incremental and eligible for procurement “existing” generation from already “installed capacity” that “has been sold to third parties.” When asked why electricity Celgar had been selling to Northpoint and FortisBC under existing and terminable contracts did not qualify as “incremental generation” under the very terms of Addendum 8, Mr. Dyck responded that Addendum 8 “is not my document. This is Power Acquisition’s document.”17 Mr. Dyck thus understood that his task encompassed more than just power acquisition. He then stated that, for Celgar, he followed his own “interpretation,” one of “determining what was incremental to what had been generated.”18 This interpretation, of course, flatly is inconsistent with Addendum 8, which specifically defined “what had been generated” as eligible, incremental power as long as it had been sold to third-parties and not used for self-supply. Canada cannot claim that Celgar’s GBL-based sales prohibition is purely procurement-related when it departs from BC Hydro’s own procurement specifications. 11. Too, Canada’s contention that the prohibition on below-GBL sales to third-parties is procurement-related because it is necessary to assure BC Hydro “security of supply” is fatuous. BC Hydro’s Mr. Scouras claimed that, without the provision, a proponent could elect to sell electricity promised to BC Hydro to a third-party instead.19 But Celgar’s promise to supply 238 GWh/yr of firm electricity to BC Hydro already effectively precludes it from selling that electricity to a third-party, as 16 R-121, BC Hydro Bioenergy Call for Power (Phase 10 Addendum 8 (7 May 2008), p. 4, § 8 (emphasis added). See also Scouras First Witness Statement, ¶ 44 (explaining that the “Existing Contract” language meant that the existing contract could lawfully be terminated prior to the Commercial Operation Date in the EPA.). 17 L. Dyck, Tr. 1487:13-14. 18 L. Dyck, Tr. 1490:3-4. 19 Scouras Second Witness Statement, ¶ 8; Rejoinder, ¶ 215. - 6 -',
"Mr. Dyck, during the negotiations for Celgar's agreement with BC Hydro, was there any discussion about selling power to third parties before the agreement was finalized?\nYes, there were discussions about the possibility, but the agreement ultimately allowed Celgar to sell all its existing capacity to third parties.\nAre you saying the agreement did not restrict below-GBL sales to third parties?\nThat's correct. The final agreement did not impose any such restrictions. It focused primarily on ensuring Celgar's supply commitments to BC Hydro.\nAnd what about the changes made in November 2008 regarding those sales provisions? Are you aware of any alterations affecting third-party agreements?\nTo my knowledge, the November 2008 adjustments did not impact our ability to sell to third parties under the GBL.\nJust to clarify, are you stating that there was no modification that introduced a restriction on below-GBL sales?\nCorrect, there was no such modification in the agreement.",
"Mr. Merwin, can you clarify your understanding of the term 'normal operations' as it pertains to the agreements you had with BC Hydro?\nCertainly. At the time, I understood 'normal operations' to mean what our usual electricity production levels were, with some flexibility for unforeseen changes. We believed this would be adjusted in our agreements accordingly.\nAccording to Mr. Dyck, there was no confusion on your end regarding 'normal operations', yet you are claiming otherwise. Can you explain this discrepancy?\nI recall there was definitely some confusion on our side. We asked for further clarification on several occasions, but the responses were vague. It's possible Mr. Dyck might not remember all his conversations accurately.\nAnd when it comes to the GBL set during the 2009 EPA, would you say BC Hydro overstepped by imposing a self-supply obligation on Celgar?\nNot exactly. The self-supply obligation was something we expected as part of our arrangement with BC Hydro. It was standard procedure, and we were fully prepared to adhere to it.",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9303, 0.9251],
# [0.9303, 1.0000, 0.9489],
# [0.9251, 0.9489, 1.0000]])
Evaluation
Metrics
Information Retrieval
- Dataset:
dim_768 - Evaluated with
InformationRetrievalEvaluatorwith these parameters:{ "truncate_dim": 768 }
| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.0909 |
| cosine_accuracy@3 | 0.2273 |
| cosine_accuracy@5 | 0.2727 |
| cosine_accuracy@10 | 0.3409 |
| cosine_precision@1 | 0.0909 |
| cosine_precision@3 | 0.0758 |
| cosine_precision@5 | 0.0545 |
| cosine_precision@10 | 0.0341 |
| cosine_recall@1 | 0.0909 |
| cosine_recall@3 | 0.2273 |
| cosine_recall@5 | 0.2727 |
| cosine_recall@10 | 0.3409 |
| cosine_ndcg@10 | 0.2102 |
| cosine_mrr@10 | 0.1689 |
| cosine_map@100 | 0.1795 |
Information Retrieval
- Dataset:
dim_512 - Evaluated with
InformationRetrievalEvaluatorwith these parameters:{ "truncate_dim": 512 }
| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.0758 |
| cosine_accuracy@3 | 0.2197 |
| cosine_accuracy@5 | 0.2576 |
| cosine_accuracy@10 | 0.3409 |
| cosine_precision@1 | 0.0758 |
| cosine_precision@3 | 0.0732 |
| cosine_precision@5 | 0.0515 |
| cosine_precision@10 | 0.0341 |
| cosine_recall@1 | 0.0758 |
| cosine_recall@3 | 0.2197 |
| cosine_recall@5 | 0.2576 |
| cosine_recall@10 | 0.3409 |
| cosine_ndcg@10 | 0.1992 |
| cosine_mrr@10 | 0.1546 |
| cosine_map@100 | 0.1649 |
Information Retrieval
- Dataset:
dim_256 - Evaluated with
InformationRetrievalEvaluatorwith these parameters:{ "truncate_dim": 256 }
| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.0833 |
| cosine_accuracy@3 | 0.2121 |
| cosine_accuracy@5 | 0.2576 |
| cosine_accuracy@10 | 0.3409 |
| cosine_precision@1 | 0.0833 |
| cosine_precision@3 | 0.0707 |
| cosine_precision@5 | 0.0515 |
| cosine_precision@10 | 0.0341 |
| cosine_recall@1 | 0.0833 |
| cosine_recall@3 | 0.2121 |
| cosine_recall@5 | 0.2576 |
| cosine_recall@10 | 0.3409 |
| cosine_ndcg@10 | 0.2017 |
| cosine_mrr@10 | 0.1585 |
| cosine_map@100 | 0.1701 |
Information Retrieval
- Dataset:
dim_128 - Evaluated with
InformationRetrievalEvaluatorwith these parameters:{ "truncate_dim": 128 }
| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.0833 |
| cosine_accuracy@3 | 0.1818 |
| cosine_accuracy@5 | 0.2424 |
| cosine_accuracy@10 | 0.3106 |
| cosine_precision@1 | 0.0833 |
| cosine_precision@3 | 0.0606 |
| cosine_precision@5 | 0.0485 |
| cosine_precision@10 | 0.0311 |
| cosine_recall@1 | 0.0833 |
| cosine_recall@3 | 0.1818 |
| cosine_recall@5 | 0.2424 |
| cosine_recall@10 | 0.3106 |
| cosine_ndcg@10 | 0.1851 |
| cosine_mrr@10 | 0.1463 |
| cosine_map@100 | 0.1581 |
Information Retrieval
- Dataset:
dim_64 - Evaluated with
InformationRetrievalEvaluatorwith these parameters:{ "truncate_dim": 64 }
| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.053 |
| cosine_accuracy@3 | 0.1364 |
| cosine_accuracy@5 | 0.1818 |
| cosine_accuracy@10 | 0.2727 |
| cosine_precision@1 | 0.053 |
| cosine_precision@3 | 0.0455 |
| cosine_precision@5 | 0.0364 |
| cosine_precision@10 | 0.0273 |
| cosine_recall@1 | 0.053 |
| cosine_recall@3 | 0.1364 |
| cosine_recall@5 | 0.1818 |
| cosine_recall@10 | 0.2727 |
| cosine_ndcg@10 | 0.1503 |
| cosine_mrr@10 | 0.1127 |
| cosine_map@100 | 0.1262 |
Training Details
Training Dataset
json
- Dataset: json
- Size: 1,179 training samples
- Columns:
positiveandanchor - Approximate statistics based on the first 1000 samples:
positive anchor type string string details - min: 85 tokens
- mean: 433.39 tokens
- max: 512 tokens
- min: 117 tokens
- mean: 221.02 tokens
- max: 378 tokens
- Samples:
positive anchor COD on the 2009 EPA. Tembec and BC Hydro signed a new ESA on December 7, 2009 and the mill reached COD on the 2009 EPA in November 2009. While the mill had met other commercial and technical requirements by the time the EPA was signed in August 2009, the delayed COD was the result of a new BC court decision requiring BC Hydro and/or proponents of projects similar to Skookumchuck’s to demonstrate adequate consultation of all First Nations who may have interests in the areas of operations. BC Hydro required such evidence in order to support its filing of the EPA before the BCUC under Section 71 of the BC Utilities Commission Act. The delay in COD 57. Mr. Switlishoff describes Tembec’s 2009 EPA with BC Hydro as a To support his assertion, he points to the fact that Mr. Switlishoff ignores the reasons for this 22Can you clarify the role of the BC court decision in the delay of the mill's Commercial Operation Date in 2009?
Certainly. The delay was due to a new BC court decision that required adequate consultation with all First Nations with potential interests in the area. This was necessary for BC Hydro to support the EPA filing before the BCUC.
And what steps were involved in meeting the requirements outlined by that decision?
BC Hydro, along with project proponents like Tembec, had to demonstrate that they had consulted with First Nations. This was essential to comply with Section 71 of the BC Utilities Commission Act.
Regarding the Generation Baseline Level or GBL, how was this concept applied in the context of new generation projects?
The GBL was determined using historical generation data from existing generators. New generation projects and incremental self-generation were eligible, but the GBL served as a reference point to measure incremental generation for sale. Submissions were requi...it even constitutes a well-defined, objective standard capable of being consistently applied without discretion. The answer plainly is no. Indeed, it bears none of the indicia of an objective standard. (i) The Standard Did Not Exist In Writing At Any Relevant Time 263. The first problem is that the “current normal” was not written down anywhere at the time BC Hydro purports to have applied it, and, as demonstrated in the preceding section, has been described by BC Hydro differently at different times. Canada begins its consistent methodology argument by simply asserting a standard, without identifying any source.304 The Counter-Memorial simply references Mr. Dyck’s testimony, which, at paragraphs 44 through 46, likewise describes a standard without reference to any source. 264. The standard Mr. Dyck propounds in his testimony for this proceeding exists there and not in any contemporaneous document in existence at the time BC Hydro and the BCUC made any of the GBL determinations at issu...Mr. Smith, could you clarify the basis on which the BCUC assessed the harm to BC Hydro ratepayers in the 2009 order?
Certainly. The BCUC assessed the harm at approximately C$20 million per year, based on the submissions from BC Hydro and estimates from their staff.
But isn't it true that BC Hydro's initial assessment was C$16.7 million and the BCUC staff estimated C$12.3 million?
I believe there were discussions of higher impacts at some point, possibly in internal analyses. But the fundamental concern was the potential for unjust enrichment through arbitrage.
And regarding the GBLs, you mentioned in your testimony that Tembec provided evidence to support their claim for a GBL adjustment, correct?
Yes, Tembec had detailed internal documents substantiating their generation and consumption patterns, which were taken into account by BC Hydro.electricity supply. The self-sufficiency policy also required BC Hydro to acquire an additional 3,000 GWh of “insurance” energy (i.e., beyond what was required to meet customers’ demand) by the year 2026. 78. The self-sufficiency requirement opened up opportunities for the private sector to sell clean and renewable energy to BC Hydro through a variety of competitive processes, including two Bioenergy Calls for Power. While in practice BC Hydro (through its trading arm, Powerex) continued both to import and to export electricity, it also conducted a series of acquisition processes to purchase the rights to electricity in BC to meet the self-sufficiency requirement because it could no longer rely on the spot market to meet electricity demand (as it had under previous planning assumptions that allowed for a “market allowance” during low water years). 79. Long term contracts with IPPs and industrial self-generators put upward pressure on BC Hydro’s electricity rates, as the cost of new sup...Mr. Thompson, can you clarify BC Hydro’s policy on electricity self-sufficiency?
Certainly. BC Hydro had a policy that aimed for full self-sufficiency by 2026, including an extra 3,000 GWh as a buffer.
And did this policy affect the structuring of contracts with independent power producers?
Yes, the policy led to numerous long-term contracts with IPPs, which did indeed raise the average rates slightly because these new suppliers charged a bit more than BC Hydro's own resources.
Is it correct that Powerex, BC Hydro’s trading arm, was restricted from engaging in certain trades due to this policy?
That's right, Powerex focused primarily on international markets since domestic trading was limited to maintain self-sufficiency.
And what about the role of the Ministry of Energy and Mines in overseeing these strategic decisions?
The Ministry did oversee the major strategic directions, but they allowed considerable autonomy for BC Hydro and Powerex in terms of operational decisions. - Loss:
MatryoshkaLosswith these parameters:{ "loss": "MultipleNegativesRankingLoss", "matryoshka_dims": [ 768, 512, 256, 128, 64 ], "matryoshka_weights": [ 1, 1, 1, 1, 1 ], "n_dims_per_step": -1 }
Training Hyperparameters
Non-Default Hyperparameters
eval_strategy: epochper_device_train_batch_size: 4per_device_eval_batch_size: 4gradient_accumulation_steps: 128learning_rate: 2e-05num_train_epochs: 4lr_scheduler_type: cosinewarmup_ratio: 0.1tf32: Falseload_best_model_at_end: Truedataloader_pin_memory: Falsegradient_checkpointing: Truebatch_sampler: no_duplicates
All Hyperparameters
Click to expand
overwrite_output_dir: Falsedo_predict: Falseeval_strategy: epochprediction_loss_only: Trueper_device_train_batch_size: 4per_device_eval_batch_size: 4per_gpu_train_batch_size: Noneper_gpu_eval_batch_size: Nonegradient_accumulation_steps: 128eval_accumulation_steps: Nonelearning_rate: 2e-05weight_decay: 0.0adam_beta1: 0.9adam_beta2: 0.999adam_epsilon: 1e-08max_grad_norm: 1.0num_train_epochs: 4max_steps: -1lr_scheduler_type: cosinelr_scheduler_kwargs: {}warmup_ratio: 0.1warmup_steps: 0log_level: passivelog_level_replica: warninglog_on_each_node: Truelogging_nan_inf_filter: Truesave_safetensors: Truesave_on_each_node: Falsesave_only_model: Falserestore_callback_states_from_checkpoint: Falseno_cuda: Falseuse_cpu: Falseuse_mps_device: Falseseed: 42data_seed: Nonejit_mode_eval: Falseuse_ipex: Falsebf16: Falsefp16: Falsefp16_opt_level: O1half_precision_backend: autobf16_full_eval: Falsefp16_full_eval: Falsetf32: Falselocal_rank: 0ddp_backend: Nonetpu_num_cores: Nonetpu_metrics_debug: Falsedebug: []dataloader_drop_last: Falsedataloader_num_workers: 0dataloader_prefetch_factor: Nonepast_index: -1disable_tqdm: Falseremove_unused_columns: Truelabel_names: Noneload_best_model_at_end: Trueignore_data_skip: Falsefsdp: []fsdp_min_num_params: 0fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap: Noneaccelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed: Nonelabel_smoothing_factor: 0.0optim: adamw_torchoptim_args: Noneadafactor: Falsegroup_by_length: Falselength_column_name: lengthddp_find_unused_parameters: Noneddp_bucket_cap_mb: Noneddp_broadcast_buffers: Falsedataloader_pin_memory: Falsedataloader_persistent_workers: Falseskip_memory_metrics: Trueuse_legacy_prediction_loop: Falsepush_to_hub: Falseresume_from_checkpoint: Nonehub_model_id: Nonehub_strategy: every_savehub_private_repo: Falsehub_always_push: Falsegradient_checkpointing: Truegradient_checkpointing_kwargs: Noneinclude_inputs_for_metrics: Falseeval_do_concat_batches: Truefp16_backend: autopush_to_hub_model_id: Nonepush_to_hub_organization: Nonemp_parameters:auto_find_batch_size: Falsefull_determinism: Falsetorchdynamo: Noneray_scope: lastddp_timeout: 1800torch_compile: Falsetorch_compile_backend: Nonetorch_compile_mode: Nonedispatch_batches: Nonesplit_batches: Noneinclude_tokens_per_second: Falseinclude_num_input_tokens_seen: Falseneftune_noise_alpha: Noneoptim_target_modules: Nonebatch_eval_metrics: Falseprompts: Nonebatch_sampler: no_duplicatesmulti_dataset_batch_sampler: proportionalrouter_mapping: {}learning_rate_mapping: {}
Training Logs
| Epoch | Step | dim_768_cosine_ndcg@10 | dim_512_cosine_ndcg@10 | dim_256_cosine_ndcg@10 | dim_128_cosine_ndcg@10 | dim_64_cosine_ndcg@10 |
|---|---|---|---|---|---|---|
| 0.8678 | 2 | 0.1660 | 0.1608 | 0.1488 | 0.1316 | 0.1352 |
| 1.7356 | 4 | 0.1961 | 0.1904 | 0.1859 | 0.1645 | 0.1545 |
| 2.6034 | 6 | 0.2084 | 0.1979 | 0.1975 | 0.1817 | 0.1585 |
| 3.4712 | 8 | 0.2102 | 0.1992 | 0.2017 | 0.1851 | 0.1503 |
- The bold row denotes the saved checkpoint.
Framework Versions
- Python: 3.10.9
- Sentence Transformers: 5.1.0
- Transformers: 4.41.2
- PyTorch: 2.1.2
- Accelerate: 1.7.0
- Datasets: 4.0.0
- Tokenizers: 0.19.1
Citation
BibTeX
Sentence Transformers
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
MatryoshkaLoss
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
MultipleNegativesRankingLoss
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
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},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}