INDEX

Explanations

words related to negative behaviors or actions, especially harassment

oai_token-act-pair · gpt-3.5-turbo

instances of the term "harassment" and related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

éĹĺ

-0.91

stanbul

-0.79

rient

-0.77

zyme

-0.77

ACTED

-0.74

ECD

-0.73

iets

-0.71

inet

-0.69

arch

-0.68

kos

-0.68

POSITIVE LOGITS

 harassment

1.00

 harass

0.96

 harassing

0.93

 accus

0.89

 stalking

0.82

 tactics

0.81

 harassed

0.80

assment

0.78

 leveled

0.72

 inflic

0.72

Activations Density 0.030%

words related to negative behaviors or actions, especially harassment

instances of the term "harassment" and related contexts

No Comments

No Known Activations