INDEX

Explanations

phrases related to bypassing security measures

oai_token-act-pair · gpt-3.5-turbo

terms related to evading or bypassing rules and regulations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sad

-0.67

uesday

-0.67

killer

-0.66

soon

-0.66

mad

-0.64

iaries

-0.64

 breaths

-0.64

oom

-0.63

asp

-0.62

aeus

-0.62

POSITIVE LOGITS

 detection

1.20

 prohibitions

0.85

 FOIA

0.83

 restrictions

0.81

 regulations

0.81

 censorship

0.81

 bounds

0.79

 Regulation

0.78

 pesky

0.78

 Detection

0.77

Activations Density 0.126%

phrases related to bypassing security measures

terms related to evading or bypassing rules and regulations

No Comments

No Known Activations