INDEX

Explanations

instances where an action or situation is being emphasized or highlighted

oai_token-act-pair · gpt-3.5-turbo

terms related to the concept of being overt or covert

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

nesota

-0.84

UTION

-0.75

 Occupations

-0.74

ERAL

-0.68

 Caps

-0.66

«ĺ

-0.66

©¶æ

-0.65

lihood

-0.65

halla

-0.62

 dearly

-0.62

POSITIVE LOGITS

orial

1.08

uned

1.05

aken

1.03

itled

0.89

uning

0.89

orical

0.88

imes

0.87

aking

0.87

empt

0.84

urned

0.83

Activations Density 0.026%

instances where an action or situation is being emphasized or highlighted

terms related to the concept of being overt or covert

No Comments

No Known Activations