Merge branch 'master' into multilingual

tests/testdata/lambada_mt_en-v0-res.json

+{"results": {"lambada_mt_en": {"acc": 0.0, "acc_stderr": 0.0, "ppl": 1.6479047769869253, "ppl_stderr": 0.006497321146240192}}, "versions": {"lambada_mt_en": 0}}
\ No newline at end of file

tests/testdata/lambada_mt_es-v0-loglikelihood

+4a88f4b316c72fe0396c382d6cbb33568ac4d0ad225150d3536635c085359fc9
\ No newline at end of file

tests/testdata/lambada_mt_es-v0-res.json

+{"results": {"lambada_mt_es": {"acc": 0.0, "acc_stderr": 0.0, "ppl": 1.6479047769869253, "ppl_stderr": 0.006497321146240192}}, "versions": {"lambada_mt_es": 0}}
\ No newline at end of file

tests/testdata/lambada_mt_fr-v0-loglikelihood

+5d16f4a0c51dc6d7b6df2ebeba2bbfa51e700b843779b559b3d90183d7b02a11
\ No newline at end of file

tests/testdata/lambada_mt_fr-v0-res.json

+{"results": {"lambada_mt_fr": {"acc": 0.0, "acc_stderr": 0.0, "ppl": 1.6479047769869253, "ppl_stderr": 0.006497321146240192}}, "versions": {"lambada_mt_fr": 0}}
\ No newline at end of file

tests/testdata/lambada_mt_it-v0-loglikelihood

+fd87c6c5cf4e0499c5f9f80e5bd7ee6a4f3d2991902a0cc3ec9e6eaf22d6760a
\ No newline at end of file

tests/testdata/lambada_mt_it-v0-res.json

+{"results": {"lambada_mt_it": {"acc": 0.0, "acc_stderr": 0.0, "ppl": 1.6479047769869253, "ppl_stderr": 0.006497321146240192}}, "versions": {"lambada_mt_it": 0}}
\ No newline at end of file

tests/testdata/mc_taco-v0-loglikelihood

+1811808ef05afd5f30ffc3471622a3dd7a1b681b17a2f7616695ad6b2a45943c
\ No newline at end of file

tests/testdata/mc_taco-v0-res.json

+{"results": {"mc_taco": {"em": 0.07732732732732733, "f1": 0.41600515965511614}}, "versions": {"mc_taco": 0}}
\ No newline at end of file

tests/testdata/mutual-v0-loglikelihood

+f759213a28f0412510bf1a24c9cab0dae64bdee902d42a26225295445e7779db
\ No newline at end of file

tests/testdata/mutual-v0-res.json

+{"results": {"mutual": {"mrr": 0.5023513920240772, "mrr_stderr": 0.009501864812936679, "r@1": 0.22573363431151242, "r@1_stderr": 0.014053085820407457, "r@2": 0.4221218961625282, "r@2_stderr": 0.016602191705517556}}, "versions": {"mutual": 0}}
\ No newline at end of file

tests/testdata/mutual-v1-loglikelihood

+f759213a28f0412510bf1a24c9cab0dae64bdee902d42a26225295445e7779db
\ No newline at end of file

tests/testdata/mutual-v1-res.json

+{"results": {"mutual": {"mrr": 0.5023513920240772, "mrr_stderr": 0.009501864812936679, "r@1": 0.22460496613995484, "r@1_stderr": 0.014028122493992806, "r@2": 0.4706546275395034, "r@2_stderr": 0.016778343895001414}}, "versions": {"mutual": 1}}
\ No newline at end of file

tests/testdata/mutual_plus-v0-loglikelihood

+b846bb9db109535f59a93d1ce340cf09f68bdf4fed5b8decd168784220fe07fa
\ No newline at end of file

tests/testdata/mutual_plus-v0-res.json

+{"results": {"mutual_plus": {"mrr": 0.5275583145221953, "mrr_stderr": 0.009940894824430708, "r@1": 0.2595936794582393, "r@1_stderr": 0.014737047402750955, "r@2": 0.45372460496614, "r@2_stderr": 0.01673517854461967}}, "versions": {"mutual_plus": 0}}
\ No newline at end of file

tests/testdata/mutual_plus-v1-loglikelihood

+b846bb9db109535f59a93d1ce340cf09f68bdf4fed5b8decd168784220fe07fa
\ No newline at end of file

tests/testdata/mutual_plus-v1-res.json

+{"results": {"mutual_plus": {"mrr": 0.5275583145221953, "mrr_stderr": 0.009940894824430708, "r@1": 0.26297968397291194, "r@1_stderr": 0.01479889176605113, "r@2": 0.5, "r@2_stderr": 0.01680731613632036}}, "versions": {"mutual_plus": 1}}
\ No newline at end of file

tests/testdata/pile_arxiv-v1-loglikelihood_rolling

+814f9954e44368559602c00f7e85fa3971acdfd0315f508ec7df6318a79c55ec
\ No newline at end of file

tests/testdata/pile_arxiv-v1-res.json

+{"results": {"pile_arxiv": {"bits_per_byte": 1.55095665856779e-05, "byte_perplexity": 1.0000107504701365, "word_perplexity": 1.0000819333090385}}, "versions": {"pile_arxiv": 1}}
\ No newline at end of file

tests/testdata/pile_bookcorpus2-v1-loglikelihood_rolling

+5c17ddfebeab8c41dabadb6fc216ceda91e3fe5dc95aaf1b2c843d7f11828b03
\ No newline at end of file