INDEX

Explanations

instances where a request for user verification against being a robot is made

oai_token-act-pair · gpt-3.5-turbo

second-person pronouns and related verification language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kefeller

-0.81

 Wonderland

-0.68

zhen

-0.67

uries

-0.66

minecraft

-0.65

Hunt

-0.64

models

-0.64

NetMessage

-0.64

atown

-0.63

igans

-0.61

POSITIVE LOGITS

RS

0.74

iciency

0.69

 verify

0.66

ve

0.65

nos

0.62

 neutrality

0.61

SA

0.61

email

0.60

're

0.59

asses

0.59

Activations Density 0.032%

instances where a request for user verification against being a robot is made

second-person pronouns and related verification language

No Comments

No Known Activations