INDEX

Explanations

the word "anywhere" with a high activation

oai_token-act-pair · gpt-3.5-turbo

repeated references to the word "anywhere."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

asting

-0.67

asts

-0.64

othy

-0.63

seless

-0.63

ylene

-0.63

uers

-0.63

apo

-0.61

rc

-0.60

 Reef

-0.60

ulously

-0.60

POSITIVE LOGITS

 else

1.23

abouts

1.13

Else

1.01

 Else

0.99

where

0.86

else

0.82

 imaginable

0.82

ħĭ

0.79

sterdam

0.77

ĪĴ

0.75

Activations Density 0.013%

the word "anywhere" with a high activation

repeated references to the word "anywhere."

No Comments

No Known Activations