INDEX

Explanations

words related to frustration or annoyance

oai_token-act-pair · gpt-3.5-turbo

emotional expressions of frustration and discontent

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anyl

-0.62

ujah

-0.54

allo

-0.54

ROM

-0.50

hu

-0.50

hiba

-0.50

 parks

-0.49

aro

-0.46

QR

-0.46

agna

-0.45

POSITIVE LOGITS

terness

0.80

 suspic

0.75

 mathemat

0.72

 caut

0.67

ª

0.66

ĵĺ

0.65

ï¸

0.64

etheless

0.64

 agre

0.61

reditary

0.61

Activations Density 0.935%

words related to frustration or annoyance

emotional expressions of frustration and discontent

No Comments

No Known Activations