INDEX

Explanations

phrases indicating rejection and disapproval

oai_token-act-pair · gpt-3.5-turbo

instances of the word "reject" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ivil

-0.72

ortal

-0.69

tesy

-0.68

isd

-0.68

imeo

-0.66

activated

-0.66

ammy

-0.66

uyomi

-0.64

RH

-0.64

ttp

-0.64

POSITIVE LOGITS

 reject

0.87

 rejection

0.84

 validation

0.81

kus

0.79

ively

0.76

furt

0.70

 outright

0.69

atos

0.68

uously

0.66

pport

0.65

Activations Density 0.009%

phrases indicating rejection and disapproval

instances of the word "reject" and its variations

No Comments

No Known Activations