Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- commit c13dac8e2974599969b18a4a08b38d08db6ba019
- Author: Shyam Pather <shyam.pather@gmail.com>
- Date: Thu Sep 21 09:10:00 2023 -0700
- Run the attention weights experiments on all layers/heads
- Got through all layers/heads on s_len=7 and through block 0/head 4 on
- the s_len=256 run. Not committing the binaries for the s_len=256 run
- because the file sizes are too large. Will work on a solution to split
- them.
- diff --git a/nbs/attn_weight_results/block0_head0_slen7_data.pt b/nbs/attn_weight_results/block0_head0_slen7_data.pt
- new file mode 100644
- index 0000000..aa97a5a
- --- /dev/null
- b/nbs/attn_weight_results/block0_head0_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:d376969de8ffee82408dc71debe1b9fc8700f7b31488800c17f59754a4de488a
- size 14421915
- diff --git a/nbs/attn_weight_results/block0_head1_slen7_data.pt b/nbs/attn_weight_results/block0_head1_slen7_data.pt
- new file mode 100644
- index 0000000..a3c03f8
- --- /dev/null
- b/nbs/attn_weight_results/block0_head1_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:42cd74adde989b1d8764ba03fa4d8616b3cc5428fd846079170c5310b00e654f
- size 14421915
- diff --git a/nbs/attn_weight_results/block0_head2_slen7_data.pt b/nbs/attn_weight_results/block0_head2_slen7_data.pt
- new file mode 100644
- index 0000000..5250503
- --- /dev/null
- b/nbs/attn_weight_results/block0_head2_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:593d89e7bbf7bfddf2eba0e383c6be549ba06cc0944892cc5985cd36c178a193
- size 14421915
- diff --git a/nbs/attn_weight_results/block0_head3_slen7_data.pt b/nbs/attn_weight_results/block0_head3_slen7_data.pt
- new file mode 100644
- index 0000000..d078b41
- --- /dev/null
- b/nbs/attn_weight_results/block0_head3_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:ae446b615c9fb213f71d646f45df0b1c14a57a7c394f7d5b30fef9ca73c63a3b
- size 14421915
- diff --git a/nbs/attn_weight_results/block0_head4_slen7_data.pt b/nbs/attn_weight_results/block0_head4_slen7_data.pt
- new file mode 100644
- index 0000000..ed2bb16
- --- /dev/null
- b/nbs/attn_weight_results/block0_head4_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:5b87969f59878e4260de8759d3213c08432fbf36a712a978b837e584bb91bf25
- size 14421915
- diff --git a/nbs/attn_weight_results/block0_head5_slen7_data.pt b/nbs/attn_weight_results/block0_head5_slen7_data.pt
- new file mode 100644
- index 0000000..4b6425b
- --- /dev/null
- b/nbs/attn_weight_results/block0_head5_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:5dd6f7b14c913605f1b3112f578e0cffcb36610816f1b5e2b5415c68e8524ff8
- size 14421915
- diff --git a/nbs/attn_weight_results/block1_head1_slen7_data.pt b/nbs/attn_weight_results/block1_head1_slen7_data.pt
- new file mode 100644
- index 0000000..e3784b5
- --- /dev/null
- b/nbs/attn_weight_results/block1_head1_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:de933a23375f409ef142871a7bd90ab66ba2349fbe68d3ec658aae52364a002f
- size 14421915
- diff --git a/nbs/attn_weight_results/block1_head2_slen7_data.pt b/nbs/attn_weight_results/block1_head2_slen7_data.pt
- new file mode 100644
- index 0000000..a25d04f
- --- /dev/null
- b/nbs/attn_weight_results/block1_head2_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:7e8dcccacbc67ea3cd588482f803e03046b3e27d3433e9906e9d0b80a8856e2e
- size 14421915
- diff --git a/nbs/attn_weight_results/block1_head3_slen7_data.pt b/nbs/attn_weight_results/block1_head3_slen7_data.pt
- new file mode 100644
- index 0000000..057a8c1
- --- /dev/null
- b/nbs/attn_weight_results/block1_head3_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:7b43c4aaae0b614ca2ae213c884e54e30e0b613465af6312b70518737dfeccb2
- size 14421915
- diff --git a/nbs/attn_weight_results/block1_head4_slen7_data.pt b/nbs/attn_weight_results/block1_head4_slen7_data.pt
- new file mode 100644
- index 0000000..f5db456
- --- /dev/null
- b/nbs/attn_weight_results/block1_head4_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:d154167d9d7303d775ddc22cc9b14e4e9f623bb198b510fab08ef053c6be8b41
- size 14421915
- diff --git a/nbs/attn_weight_results/block1_head5_slen7_data.pt b/nbs/attn_weight_results/block1_head5_slen7_data.pt
- new file mode 100644
- index 0000000..bb41687
- --- /dev/null
- b/nbs/attn_weight_results/block1_head5_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:13761fec7a4f9514d171fc89f1a94e9bd10d53ecb2d46c452a08d5badaccf8b8
- size 14421915
- diff --git a/nbs/attn_weight_results/block2_head0_slen7_data.pt b/nbs/attn_weight_results/block2_head0_slen7_data.pt
- new file mode 100644
- index 0000000..cd62f6f
- --- /dev/null
- b/nbs/attn_weight_results/block2_head0_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:5b2593e8ba9356d3624b365f4dc223319f8217a9cd4bfb4660937053287b7175
- size 14421915
- diff --git a/nbs/attn_weight_results/block2_head1_slen7_data.pt b/nbs/attn_weight_results/block2_head1_slen7_data.pt
- new file mode 100644
- index 0000000..e9881f6
- --- /dev/null
- b/nbs/attn_weight_results/block2_head1_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:2502e2ed12150ccd9b38caa981e1059a07a0a764c932fd7f709717b65775e53a
- size 14421915
- diff --git a/nbs/attn_weight_results/block2_head2_slen7_data.pt b/nbs/attn_weight_results/block2_head2_slen7_data.pt
- new file mode 100644
- index 0000000..da804f5
- --- /dev/null
- b/nbs/attn_weight_results/block2_head2_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:665453d09612dc95625c3c988ddda0fe43d4944b4b3184281b61c345d5515c9a
- size 14421915
- diff --git a/nbs/attn_weight_results/block2_head3_slen7_data.pt b/nbs/attn_weight_results/block2_head3_slen7_data.pt
- new file mode 100644
- index 0000000..2fa9165
- --- /dev/null
- b/nbs/attn_weight_results/block2_head3_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:8d808f8239cdfe36012728f6a37e6f9d176b3d9d8f7fbdafe90aef28b18235f0
- size 14421915
- diff --git a/nbs/attn_weight_results/block2_head4_slen7_data.pt b/nbs/attn_weight_results/block2_head4_slen7_data.pt
- new file mode 100644
- index 0000000..1bf230f
- --- /dev/null
- b/nbs/attn_weight_results/block2_head4_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:51dbf926b11aa97619af02ecd609a0ab4bb7aa21820db2376a64f9d8c4e0dad3
- size 14421915
- diff --git a/nbs/attn_weight_results/block2_head5_slen7_data.pt b/nbs/attn_weight_results/block2_head5_slen7_data.pt
- new file mode 100644
- index 0000000..43d0a43
- --- /dev/null
- b/nbs/attn_weight_results/block2_head5_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:4bd0e67b34a96094de85a82a7757a80f7dc60bc40e586501a927ae9188fa6dc1
- size 14421915
- diff --git a/nbs/attn_weight_results/block3_head0_slen7_data.pt b/nbs/attn_weight_results/block3_head0_slen7_data.pt
- new file mode 100644
- index 0000000..c0d3478
- --- /dev/null
- b/nbs/attn_weight_results/block3_head0_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:683efc7b4e489caa8f4558cfe6b134dceb93347519ff57508e5544398e66cc0a
- size 14421915
- diff --git a/nbs/attn_weight_results/block3_head1_slen7_data.pt b/nbs/attn_weight_results/block3_head1_slen7_data.pt
- new file mode 100644
- index 0000000..b5224d3
- --- /dev/null
- b/nbs/attn_weight_results/block3_head1_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:72a2edc4044c3336285ef6fbd5997cb42c602d63d872eacf9d9da87ee30ad36b
- size 14421915
- diff --git a/nbs/attn_weight_results/block3_head2_slen7_data.pt b/nbs/attn_weight_results/block3_head2_slen7_data.pt
- new file mode 100644
- index 0000000..a00dc24
- --- /dev/null
- b/nbs/attn_weight_results/block3_head2_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:dfb0fd929904c7a479ae521a2d9c13c04327801d5db226919d29a1d56a4553f3
- size 14421915
- diff --git a/nbs/attn_weight_results/block3_head3_slen7_data.pt b/nbs/attn_weight_results/block3_head3_slen7_data.pt
- new file mode 100644
- index 0000000..aca383d
- --- /dev/null
- b/nbs/attn_weight_results/block3_head3_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:a74da8e297d3e4852f9c1c8605144211598d2bb7ede09be21d958d9441399d90
- size 14421915
- diff --git a/nbs/attn_weight_results/block3_head4_slen7_data.pt b/nbs/attn_weight_results/block3_head4_slen7_data.pt
- new file mode 100644
- index 0000000..449a74a
- --- /dev/null
- b/nbs/attn_weight_results/block3_head4_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:0f4ee0cff215c181f3cc87d8dcf61074296a73eefa892ff47258df7f4c067d64
- size 14421915
- diff --git a/nbs/attn_weight_results/block3_head5_slen7_data.pt b/nbs/attn_weight_results/block3_head5_slen7_data.pt
- new file mode 100644
- index 0000000..01a0c43
- --- /dev/null
- b/nbs/attn_weight_results/block3_head5_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:a21b98a311b5d9e54b818b714e9dd33639a4a2bd5b9e14aad9d59789d876f3b1
- size 14421915
- diff --git a/nbs/attn_weight_results/block4_head0_slen7_data.pt b/nbs/attn_weight_results/block4_head0_slen7_data.pt
- new file mode 100644
- index 0000000..58cead0
- --- /dev/null
- b/nbs/attn_weight_results/block4_head0_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:1c120ae62669f1a13f96a8a252195acab05ad05afda8084845fd9e0a4f73928e
- size 14421915
- diff --git a/nbs/attn_weight_results/block4_head1_slen7_data.pt b/nbs/attn_weight_results/block4_head1_slen7_data.pt
- new file mode 100644
- index 0000000..37ccaa7
- --- /dev/null
- b/nbs/attn_weight_results/block4_head1_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:64a38f611153cd0f3a493a27ae8451f7e203db068daec3f20a87df4e01fdf979
- size 14421915
- diff --git a/nbs/attn_weight_results/block4_head2_slen7_data.pt b/nbs/attn_weight_results/block4_head2_slen7_data.pt
- new file mode 100644
- index 0000000..8d41b9a
- --- /dev/null
- b/nbs/attn_weight_results/block4_head2_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:8f384435d1307e1da1e06095d8b867e0cf2e4f7ccd66117d77544b592f976ccc
- size 14421915
- diff --git a/nbs/attn_weight_results/block4_head3_slen7_data.pt b/nbs/attn_weight_results/block4_head3_slen7_data.pt
- new file mode 100644
- index 0000000..b69666c
- --- /dev/null
- b/nbs/attn_weight_results/block4_head3_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:6753ec74114184ad162e8148cd2dc6d9544770982de7161ecda07f524a8a2b3c
- size 14421915
- diff --git a/nbs/attn_weight_results/block4_head4_slen7_data.pt b/nbs/attn_weight_results/block4_head4_slen7_data.pt
- new file mode 100644
- index 0000000..eed6dbd
- --- /dev/null
- b/nbs/attn_weight_results/block4_head4_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:862ab2d3fff33393303ee400769ba4853b27311a58ec0755d6d83a2ea0a6ef03
- size 14421915
- diff --git a/nbs/attn_weight_results/block4_head5_slen7_data.pt b/nbs/attn_weight_results/block4_head5_slen7_data.pt
- new file mode 100644
- index 0000000..1fc399d
- --- /dev/null
- b/nbs/attn_weight_results/block4_head5_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:5c67f0625e326c1e40092f1e72ca9eb583db49e005e44da2b542eb742dfd2389
- size 14421915
- diff --git a/nbs/attn_weight_results/block5_head0_slen7_data.pt b/nbs/attn_weight_results/block5_head0_slen7_data.pt
- new file mode 100644
- index 0000000..e092b27
- --- /dev/null
- b/nbs/attn_weight_results/block5_head0_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:9ac7d665b1afe59144ff7eba36f93e2f2c3b4469dfe6608ff82d8ccda6db8660
- size 14421915
- diff --git a/nbs/attn_weight_results/block5_head1_slen7_data.pt b/nbs/attn_weight_results/block5_head1_slen7_data.pt
- new file mode 100644
- index 0000000..87cbfb8
- --- /dev/null
- b/nbs/attn_weight_results/block5_head1_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:8eb50f50a3581008186a60ef727ede61cb6fbf1cc5bbfd4f073e1340b390226d
- size 14421915
- diff --git a/nbs/attn_weight_results/block5_head2_slen7_data.pt b/nbs/attn_weight_results/block5_head2_slen7_data.pt
- new file mode 100644
- index 0000000..8c1fb88
- --- /dev/null
- b/nbs/attn_weight_results/block5_head2_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:bcf735c41f2488b29e5eef58b5d808035726871dca9755be0cde77f0dcccf07e
- size 14421915
- diff --git a/nbs/attn_weight_results/block5_head3_slen7_data.pt b/nbs/attn_weight_results/block5_head3_slen7_data.pt
- new file mode 100644
- index 0000000..6aedf10
- --- /dev/null
- b/nbs/attn_weight_results/block5_head3_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:1073af7874219c48a1d70820088d4891baa555366d3336d94d7abe8503714f21
- size 14421915
- diff --git a/nbs/attn_weight_results/block5_head4_slen7_data.pt b/nbs/attn_weight_results/block5_head4_slen7_data.pt
- new file mode 100644
- index 0000000..949bb2b
- --- /dev/null
- b/nbs/attn_weight_results/block5_head4_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:1353b908ad38868e5a27ea24d2249b731635bb1f68206f32098e90d829bc841f
- size 14421915
- diff --git a/nbs/attn_weight_results/block5_head5_slen7_data.pt b/nbs/attn_weight_results/block5_head5_slen7_data.pt
- new file mode 100644
- index 0000000..0107314
- --- /dev/null
- b/nbs/attn_weight_results/block5_head5_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:5ce804e470b9a57458a5e8975fb3be18c8b70bf57740b18ba4a3b4ac2d27bf7e
- size 14421915
- commit edf5509add7491716d849a113110a114d1e02563
- Author: Shyam Pather <shyam.pather@gmail.com>
- Date: Wed Sep 20 10:29:24 2023 -0700
- Make attention weights experiment run over all strings in the validation set
- diff --git a/nbs/attn_weight_results/block1_head0_slen7_data.pt b/nbs/attn_weight_results/block1_head0_slen7_data.pt
- new file mode 100644
- index 0000000..d090946
- --- /dev/null
- b/nbs/attn_weight_results/block1_head0_slen7_data.pt
- @@ -0,0 1,3 @@
- version https://git-lfs.github.com/spec/v1
- oid sha256:275e2400fc3d5af3a33b93619ba01acb73ac125ce385c0d6afa2ff0027a38a0f
- size 14421915
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement