INDEX

Explanations

instances of inappropriate behavior or actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "inappropriate" 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 @ 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

ership

-0.95

ynthesis

-0.87

raq

-0.83

chens

-0.80

thood

-0.80

uay

-0.79

abiding

-0.79

track

-0.79

prototype

-0.78

fighters

-0.78

POSITIVE LOGITS

 inappropriate

0.98

 improper

0.91

 behaviour

0.87

 inappropriately

0.85

 interference

0.84

 improperly

0.82

 behavior

0.81

 touching

0.80

 indisc

0.79

 undermin

0.79

Activations Density 0.020%

instances of inappropriate behavior or actions

occurrences of the word "inappropriate" and its variations

No Comments

No Known Activations