INDEX

Explanations

locations or entities followed by the word "at"

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "look at" along with various contextual elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

soever

-0.73

externalActionCode

-0.72

ON

-0.67

SPONSORED

-0.67

enthal

-0.65

emphasis

-0.63

arsh

-0.62

haw

-0.62

SUP

-0.62

sil

-0.62

POSITIVE LOGITS

how

1.00

 pictures

0.93

 photographs

0.89

 what

0.87

 photos

0.84

 trends

0.83

 whether

0.82

 screenshots

0.81

 yourselves

0.80

 alternatives

0.80

Activations Density 0.078%

locations or entities followed by the word "at"

instances of the phrase "look at" along with various contextual elements

No Comments

No Known Activations