diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..1377c29a0005cfc3d4f9fb8db67c55b41fc24e2d 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,76 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +archive/TEST_10_PERCENT_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/policy_puppetry_attack-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/policy_puppetry_attack-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/policy_puppetry_attack-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/top_1000_sample_5_percent_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +archive/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +baseline_true_prefix-Llama-3.2-1B-Instruct.db filter=lfs diff=lfs merge=lfs -text +baseline_true_prefix-Llama-3.2-3B-Instruct.db filter=lfs diff=lfs merge=lfs -text +baseline_true_prefix-Qwen2.5-1.5B-Instruct.db filter=lfs diff=lfs merge=lfs -text +baseline_true_prefix-Qwen2.5-3B-Instruct.db filter=lfs diff=lfs merge=lfs -text +business_logic-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +business_logic-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +business_logic-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +business_logic-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +defend_self_reminder_defense_virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text +virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db filter=lfs diff=lfs merge=lfs -text diff --git a/archive/TEST_10_PERCENT_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/TEST_10_PERCENT_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..4c2b4248edb05ef8678e0b1cdbc0e14f1c5ec146 --- /dev/null +++ b/archive/TEST_10_PERCENT_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b7671def9b02876620e75d8a49565193d47ecd3b4801a44a964f151b0fbd606 +size 1962946560 diff --git a/archive/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..3242ba88cc63e4deafcf10d3ee3230feda843747 --- /dev/null +++ b/archive/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9aaf29890d299f31a6eb8cb69daf52803a2124d202020e03e3454443bd612c31 +size 3158016 diff --git a/archive/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..eaa54c4c083868f12b7491fed9167a163a403bd7 --- /dev/null +++ b/archive/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cdf71d9995549b4b56033ab14c9846528332fc0e49ec112cfed2f7bab52610b5 +size 3944448 diff --git a/archive/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..86e15595e3e6b8f67fa33975b527fd0bb4663084 --- /dev/null +++ b/archive/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c8f1035008c494fe235e5674c4d679880003ddf41ad280029f859f99ca317081 +size 4993024 diff --git a/archive/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..2d5db36cdfcdf2832b40d33675679794bb4ab41e --- /dev/null +++ b/archive/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:695f42718edb24fe8ce0deb1dba746e341102b0daa109968a758999b9520b4b8 +size 3158016 diff --git a/archive/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..eb689998bb52ff5103127703694c5f4ea0e012ee --- /dev/null +++ b/archive/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f68f1233cb218d4406566e359c298237258fafdd4a4a8039cc59ac60a242c35f +size 3158016 diff --git a/archive/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..2a0cdcdb9afd659c27d34e93c559d4387d8c548e --- /dev/null +++ b/archive/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d72a00415e0b054bab33e958145b61605162a230ab21339f73959b94edf98aaa +size 3944448 diff --git a/archive/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..9c55a21b57c51998adbd7bb4e3eeebc9553f0ab1 --- /dev/null +++ b/archive/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9200c74863ba24cef8f3e9425cb3ee1b3de7fae36f7d7be42c84d091de799d6c +size 4993024 diff --git a/archive/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..48d89be44549c82fa422c959758b040c3c519147 --- /dev/null +++ b/archive/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf0be6f3fa9acc1c8383d6ae87fa3c51996b53325177c42d82b377b037d45a83 +size 3158016 diff --git a/archive/policy_puppetry_attack-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/policy_puppetry_attack-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..897ad89902cdb9d99c9061aadcfc5c8f869566b9 --- /dev/null +++ b/archive/policy_puppetry_attack-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b813dcbe8815bdc0eb71958ad9fff67986ba025ccacf8687af6315053c48ea1 +size 6041600 diff --git a/archive/policy_puppetry_attack-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/policy_puppetry_attack-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..518ebe7e69e509d896b63fa57f8272aa0cae4d72 --- /dev/null +++ b/archive/policy_puppetry_attack-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1831c6cb4f33fb92a9e500ae2f38a0f8ab45f11022523dc2209d151dbc55c9e1 +size 6041600 diff --git a/archive/policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..caf5d67011888b43752d884c04db8441d399beaa --- /dev/null +++ b/archive/policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:38f0629a60722959367a7d54c45b26f3c5550d20b0c31bda0ce68b040335cd01 +size 6828032 diff --git a/archive/policy_puppetry_attack-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/policy_puppetry_attack-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..2d57ad68a3b1a87f35209e66e3a735fa16369746 --- /dev/null +++ b/archive/policy_puppetry_attack-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c8c3d66a442e998b17dba907c6945b6eba513d6f18f43d5f8f5a8da05cc18f07 +size 6303744 diff --git a/archive/top_1000_sample_5_percent_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/top_1000_sample_5_percent_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..51dd4f6f201b694a68c0135d6628df9611a67fd5 --- /dev/null +++ b/archive/top_1000_sample_5_percent_policy_puppetry_attack-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b5b813883cf5bfed1dd94143671117093d259e33ffc78b62bd5191e154c3ad8 +size 1148727296 diff --git a/archive/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..20b1ff261a20d6e4e76391c1bfc8bc77c1bf1af1 --- /dev/null +++ b/archive/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2282dcfe86a980be8c9b2ab4c6c6e7809e3dc078c665a2db5c9b72f956f152b +size 1585152 diff --git a/archive/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..3347aa0715024dc1a969ebe6baf796803e483992 --- /dev/null +++ b/archive/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:baeb31b361c27bd0b16f97d726b963a45c60c07d9345ecb1f7ef2c4d77ae29ab +size 3158016 diff --git a/archive/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..7b274d2dcf95903eaca983362965e522ebe417b4 --- /dev/null +++ b/archive/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4691535336c006fdf0621f3cc89269d7d233f3906dd1b967e83a7a237da37842 +size 2633728 diff --git a/archive/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..c80cd13cec36237025c6262559f74f496cca856f --- /dev/null +++ b/archive/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc77c3a60e6481f13d96ea73e878115595a8204bc8750894b0a7c27c51532eb7 +size 2895872 diff --git a/archive/true_prefix_attack_name.db b/archive/true_prefix_attack_name.db new file mode 100644 index 0000000000000000000000000000000000000000..6acffabf6c765d6e59620c252ff6875aa6efb0af Binary files /dev/null and b/archive/true_prefix_attack_name.db differ diff --git a/archive/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..e42df1dab522b9538f67bca99b47af9651a1fdd5 --- /dev/null +++ b/archive/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a0238de7cf16d8aef3c54525cbb4865800e4c4d965580d59028d86655cab873 +size 4468736 diff --git a/archive/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..a24cc7f62ad1bc1bb5b9e1c562edd948c1501612 --- /dev/null +++ b/archive/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:012613226c11aae479f2d6661d5f8444bf79c92d39369aa671d3232e07ba39f3 +size 3682304 diff --git a/archive/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/archive/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..c682df4583ad140b5badfe8e688cb61078b2f620 --- /dev/null +++ b/archive/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea4dd088e86b1c443f7ee786b8d6b6961d02553c41ac8defcfd7441739392415 +size 4206592 diff --git a/baseline_true_prefix-Llama-3.2-1B-Instruct.db b/baseline_true_prefix-Llama-3.2-1B-Instruct.db new file mode 100644 index 0000000000000000000000000000000000000000..98ffda4382ae32bf8f209f66266d39c76a8732b6 --- /dev/null +++ b/baseline_true_prefix-Llama-3.2-1B-Instruct.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37ca85f035ab6bae3ecb490aecc5110cad3428f987f5fc220e55b83cfae6209b +size 3420160 diff --git a/baseline_true_prefix-Llama-3.2-3B-Instruct.db b/baseline_true_prefix-Llama-3.2-3B-Instruct.db new file mode 100644 index 0000000000000000000000000000000000000000..fb7cb82acf4c4a4f0fbd22b53ff4fdb89cd3680b --- /dev/null +++ b/baseline_true_prefix-Llama-3.2-3B-Instruct.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ad3ce1453b3e3199528a326403481042af4dcce0a822e9b830795dcee06030d +size 2895872 diff --git a/baseline_true_prefix-Qwen2.5-1.5B-Instruct.db b/baseline_true_prefix-Qwen2.5-1.5B-Instruct.db new file mode 100644 index 0000000000000000000000000000000000000000..0f52efc59d7ca18ba92ee7c468dcde6b13d23550 --- /dev/null +++ b/baseline_true_prefix-Qwen2.5-1.5B-Instruct.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ac1861a31a7e54e346c2f183eb2c0c77294bc455699b5277c9815ee333429a1 +size 1323008 diff --git a/baseline_true_prefix-Qwen2.5-3B-Instruct.db b/baseline_true_prefix-Qwen2.5-3B-Instruct.db new file mode 100644 index 0000000000000000000000000000000000000000..08d5728eeef4c5738bd5ba6612ea8a1874c3f24f --- /dev/null +++ b/baseline_true_prefix-Qwen2.5-3B-Instruct.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd510f010fbe3b375683241f63095449725f3c11026e4145d44af6cfafe0b526 +size 1323008 diff --git a/business_logic-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/business_logic-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..70d04f5a9f0c55d47be08c045704ab0e17e3d29c --- /dev/null +++ b/business_logic-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2149e5e726a44a3ed53f82084d9502694896ab4b0a7a8b294d4c80800d14218 +size 2371584 diff --git a/business_logic-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/business_logic-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..a7b3f091c0e330ee8a8716e4d4fe7f866acaa6c7 --- /dev/null +++ b/business_logic-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:58f935ba30f18d87903b175b71676fe5d6bb2e36d554c7108593d05c77fd6f76 +size 1585152 diff --git a/business_logic-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/business_logic-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..b15c454c355bd62d19b07ae13ff3fc0ce95dea02 --- /dev/null +++ b/business_logic-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ff455ad605428f1b40d85da1b671190f1d4c7bd01e35abfcee97a4dae2b1d93 +size 2371584 diff --git a/business_logic-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/business_logic-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..6474c58cee074b2d1032f480798d606c9f0138a0 --- /dev/null +++ b/business_logic-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9df17afd15131ff20f76588e196644aa3addc12e5e3f22b99205c3b5196555b9 +size 2371584 diff --git a/defend_self_reminder_defense-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..3aa806fd0a7642054269f039c162788ff87b8b82 --- /dev/null +++ b/defend_self_reminder_defense-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7d419dd25491005739c199f5d9025254380a73948bee56b75558665854faba4 +size 2633728 diff --git a/defend_self_reminder_defense-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..80767719f6ac7a83cb3b73bafda756e73d600960 --- /dev/null +++ b/defend_self_reminder_defense-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:894babff5104b071e8a087df1958a86947218edc40856c3d9bac37888562a132 +size 2633728 diff --git a/defend_self_reminder_defense-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..19553cfcd548d6ec9410924e05bc866a0e2952ac --- /dev/null +++ b/defend_self_reminder_defense-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2203b2f5adb5fad13468c3277c2cc700b6e3185349063f8914e689558789c55 +size 1847296 diff --git a/defend_self_reminder_defense-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..46a6b1ac95625a00afcc3d1f7e26c9d4574f29ab --- /dev/null +++ b/defend_self_reminder_defense-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:edc71218369bc197bc2938f271473c1f330edc1cb49a549c4ce2cdee21184af7 +size 2895872 diff --git a/defend_self_reminder_defense_fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..9cedc21a22284dba3f0aebaead1ff50c7e97d4e5 --- /dev/null +++ b/defend_self_reminder_defense_fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89031b097e8bc19a6e65e9863a5e62997cad86e4aa8ad361ae52bc7ebbc0c53d +size 2895872 diff --git a/defend_self_reminder_defense_fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..71034ccdd9173969ceacc5c3852448b62e7043fa --- /dev/null +++ b/defend_self_reminder_defense_fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d9d8d7240b6b3a2fb8fb5b2c8cfe1d14822533efca1560383bc7cf7a6a4bcb0 +size 2895872 diff --git a/defend_self_reminder_defense_fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..85457e6f3d2d2c3c869bdf50f0803a2953fcfc00 --- /dev/null +++ b/defend_self_reminder_defense_fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85b1332cd49c36c3161dbfd13909d6726213b0e24f2ef4fee925db32470d589c +size 4730880 diff --git a/defend_self_reminder_defense_fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..105c7de306ddc5954e74ecf2ba7207bed182eb1d --- /dev/null +++ b/defend_self_reminder_defense_fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f68aa78bc9e55168c212aa903c65f64add77dd3333cc5935156e2497ae9f276d +size 4730880 diff --git a/defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..56d2c52e4e76cd364d880a06965719954f27a1c9 --- /dev/null +++ b/defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07a304feddd25b6c013f4831d0d2b0b74d7a52d24c992e9d03d9d1ddf93335f2 +size 3682304 diff --git a/defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..6facdcf6c4cf49a7eb2ad9f911ae9169ab25002e --- /dev/null +++ b/defend_self_reminder_defense_ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47d41941ff105ba40ae96dce08b807506095b2e3acdae2df1b62f1d8a312dff4 +size 3420160 diff --git a/defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..67d92fa489a2580b8a1231a79e0148478ff8d3ab --- /dev/null +++ b/defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2b78687d24ce45266a4ab99ff14f42470705206e21c446f2ad0d6c0a0ec6c78 +size 4993024 diff --git a/defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..adb3d07115dd48404cc56e5d1309b96e2cd69fbb --- /dev/null +++ b/defend_self_reminder_defense_ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9eb6f884698ca0af28bb106f41ede129a73945fb9bac1903204728c875c50f17 +size 5255168 diff --git a/defend_self_reminder_defense_policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..6967169f4998ef6cd721bb38dcfd34f1ff25b695 --- /dev/null +++ b/defend_self_reminder_defense_policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee94ee8ba5fad7bbd1659f76213b54caee8b3b093e95467318856c5015afc0a3 +size 5779456 diff --git a/defend_self_reminder_defense_policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..10d2cd0bd3873d4a9af2b140373872d7f2ec9ad1 --- /dev/null +++ b/defend_self_reminder_defense_policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b721b90388cec81874d412b709c7edba2cbd180bbca80ed3b83b5e7fedfcb86 +size 5779456 diff --git a/defend_self_reminder_defense_policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..9c386d5bb71ddff955286f6056dfffaf8b0b5cae --- /dev/null +++ b/defend_self_reminder_defense_policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:344b2cbb0e520ae83a9286bf9b902265863377dab4e44b845738fc1ef2e12119 +size 7614464 diff --git a/defend_self_reminder_defense_policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..1d45aeef0aaa99b83372bba2e0a11c203679d1af --- /dev/null +++ b/defend_self_reminder_defense_policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6642e64be31df1a73ca51c6b9dce3c0b8ae59400f4fa022fc1ae27f7e6f0826 +size 6565888 diff --git a/defend_self_reminder_defense_true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..3c9ce61ec127d3dc639305fafe630f26218761b4 --- /dev/null +++ b/defend_self_reminder_defense_true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f785a861e394dc1330ac54c9aad0397d1ea572aa87c3cc8a51284cdf8e311e80 +size 1323008 diff --git a/defend_self_reminder_defense_true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..11a69a88b2d0db332b1fb7691e345bd7731ae3ee --- /dev/null +++ b/defend_self_reminder_defense_true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74693fe5759497714291d20c71681bc335fa85dac271a829c801bf05b9823639 +size 2633728 diff --git a/defend_self_reminder_defense_true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..aad32a5de3c23b0a5c49147476bb0af9c0e9aebb --- /dev/null +++ b/defend_self_reminder_defense_true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:065f10599be43728d29969f9237eeed958b829b51d76959ed4dfc78ee171125d +size 1847296 diff --git a/defend_self_reminder_defense_true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..194b5fcd6d12d3cd036520ac6a6803f3682e75d8 --- /dev/null +++ b/defend_self_reminder_defense_true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d632b50e7113abd9ac787e57e6d97fe209bc53783d3a0a64e7b97aa8fb56eb9d +size 1323008 diff --git a/defend_self_reminder_defense_virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..f810d9b7772976ed7c2c41243805c072350ab048 --- /dev/null +++ b/defend_self_reminder_defense_virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0739dd36209674ab34f837880cb7354d0256069f4df22bbe408d4b139269a035 +size 2109440 diff --git a/defend_self_reminder_defense_virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..af9d18cac112fc715914355f0617ff464ba014cc --- /dev/null +++ b/defend_self_reminder_defense_virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:349de47c5c5baffb84d8816bb8456ae8258ac085c441aa2f454a0e64e653ba15 +size 1847296 diff --git a/defend_self_reminder_defense_virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..083663342ab3dd69ddec83e289ed0f08656c2a23 --- /dev/null +++ b/defend_self_reminder_defense_virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5a473c4a574b553b74c3af250b6fe149c11259613f2588f4ee89dab14b55909 +size 1585152 diff --git a/defend_self_reminder_defense_virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/defend_self_reminder_defense_virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..1781e404c8e3c83270126c94c92545897c092819 --- /dev/null +++ b/defend_self_reminder_defense_virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46fd7cf8f4abbd43c04e74c604f051d62d310cf24edb0cece693cb93077c59e5 +size 2109440 diff --git a/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..b0fdb3370757e08303be2686cc2f03b952295b27 --- /dev/null +++ b/fake_completion-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:82d4c7f72f0540e99edb3f962b54bf9b06e5922d1590745d427b772ad5d1546a +size 2633728 diff --git a/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..67441384ceac64d5a013fb570ece88b49a8b9bdd --- /dev/null +++ b/fake_completion-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:484fd703feb8bd0fd333c2343e6541774ffa7ca2a1ad1f93e60bd26e2c652a97 +size 3158016 diff --git a/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..a5d9a2dfa043e46ae82d43caf88f44a1986f6329 --- /dev/null +++ b/fake_completion-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3992e02c990971448c52da2a889b1d1f7f23c28a8d8ea853a7ff6f1d9d32ddc3 +size 3944448 diff --git a/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..5fb4a24ecd2200b1c8617ae5e2da15a9644ff836 --- /dev/null +++ b/fake_completion-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5338bd31f9b4f4e8a6908b92f548aca6238b08f4262369cdd48b6c8df6dea570 +size 3682304 diff --git a/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..aa4f2a126e5342f47c83d9e7d1fab9296757af86 --- /dev/null +++ b/ignore_all_previous_instructions-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8b8dcf135ef63ad1e2ca7fae95c71c99a4d7e0283e5788ce59a77b80172c5d1 +size 2633728 diff --git a/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..295659d4eb7f87f0e3ecbe05c0566796311ac5b6 --- /dev/null +++ b/ignore_all_previous_instructions-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:363c29ff75799eb5b174380961e04650df59eea138c3e104eaf4f1ac951a6492 +size 3158016 diff --git a/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..45a69032ad6eedb44692c0bd19604e0da65fd61b --- /dev/null +++ b/ignore_all_previous_instructions-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7191effb746aa029e7d5e932931c5cfd2ea967ea5bdad7aa7e2fc558f4a6f6a4 +size 3944448 diff --git a/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..4a42bc909ac60acac4b61ad542ac18e767caf159 --- /dev/null +++ b/ignore_all_previous_instructions-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce06dfcf236a2c6d477b34b5941f338fdf51f36c77fca08a0d2206820db7b76a +size 3682304 diff --git a/policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..819950a249156dd49befeb9efdc79d7f6b45ab6d --- /dev/null +++ b/policy_puppetry-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e2022d8c0d0b8a059f62960b5a4a6dcf42b4dab2116bd52dd872259f126d871 +size 4468736 diff --git a/policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..f8934c52aba0c5cb770b3c725bdc36c5b6613f89 --- /dev/null +++ b/policy_puppetry-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c7401e34ae299e420f73c9558c00737f4fc4e659a9416919fe66eee3f76e036 +size 4468736 diff --git a/policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..c454e61a7143ca4c89650c59a7c661d7f78501cc --- /dev/null +++ b/policy_puppetry-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97efb8b7b51ec1bb90c4220d1d57234f64fbb0eb7bd38aa83ca3fe5e27526c72 +size 6828032 diff --git a/policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..cfdeafe3b6782338e1ca999b6a2b6f426da54ac1 --- /dev/null +++ b/policy_puppetry-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bac18d96a4912458d2cb76ee7ab85434a4473b6b9838076f51c11093bc757162 +size 6041600 diff --git a/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..5a66d585f128c70550235f8ff017f6d89fc42ce3 --- /dev/null +++ b/true_prefix-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ae0b865b70e89ffa855c8ec215c1b99d356d866a85ef55868eeb04716923d90 +size 1323008 diff --git a/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..bc2809bcca1fc8112ad6da606f393af8b32906ca --- /dev/null +++ b/true_prefix-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:acd0a0f35da472e0c5312a3a36dd7ef1808f816c2456424bdc602d84a3fcc0fb +size 2633728 diff --git a/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..b637bb704bc1cab2143ce486d4e1103ca2a98de8 --- /dev/null +++ b/true_prefix-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb77e5697c50799f2c0a47529a8f37128416587503402c6b34e8a4b45da974f +size 1323008 diff --git a/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..6f3609753bddfc730be225df72b6c8188f6d5246 --- /dev/null +++ b/true_prefix-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f7ce3ab21a1daf3b282f8581cc226f4c60110bad2dc7d3c70ffa9857fea4434 +size 2109440 diff --git a/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..e58bfba833988221c1475a1a456c3cd240f79b55 --- /dev/null +++ b/virtualization-Llama-3.2-1B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ba53303b92470286ab2d775d6cb0ffb84a9c94dc42fbef781130e8df9284ce8 +size 1585152 diff --git a/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..ef6aca27ea03a92915a5aa48af7572c35c5e4735 --- /dev/null +++ b/virtualization-Llama-3.2-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bcc45b6f54501244b5ce1ca34443043fbe109ea7a5271ae5b8c2360c9bfcc5c4 +size 2895872 diff --git a/virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..35ec2c4dfe48756964a48a5717a7db51927d6236 --- /dev/null +++ b/virtualization-Qwen2.5-1.5B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4638d286a410cc8306a3b3edd143bf09d89c2a0136e8f2813b936a700184b8bf +size 3420160 diff --git a/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db b/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db new file mode 100644 index 0000000000000000000000000000000000000000..1f6265422ee80cc39eb2f5dbf8a2da873cd06a9a --- /dev/null +++ b/virtualization-Qwen2.5-3B-Instruct-kivedu-unsloth-defaults-fake-pii-349.db @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e794ba89147579bcdc079cfb2513a69741161ebed40a7f6d6848ce9809473f27 +size 1585152