INDEX

Explanations

phrases related to admitting something or self-awareness

oai_token-act-pair · gpt-3.5-turbo

expressions of acknowledgment or admission of difficult truths

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ItemTracker

-0.81

srf

-0.76

 Unloaded

-0.66

ammy

-0.64

zzi

-0.62

ibaba

-0.62

 attendant

-0.61

ums

-0.61

 quint

-0.61

 Nanto

-0.60

POSITIVE LOGITS

enance

0.94

cliffe

0.86

ively

0.84

uate

0.84

ible

0.81

lled

0.75

 anything

0.73

ibly

0.73

able

0.71

rist

0.70

Activations Density 0.270%

phrases related to admitting something or self-awareness

expressions of acknowledgment or admission of difficult truths

No Comments

No Known Activations