INDEX

Explanations

words associated with attracting or persuading someone to do something

oai_token-act-pair · gpt-3.5-turbo

terms related to attracting or drawing in individuals

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

regulation

-0.77

undred

-0.71

oret

-0.69

 PRESIDENT

-0.69

actions

-0.68

arna

-0.67

sections

-0.67

iop

-0.66

endif

-0.65

airs

-0.64

POSITIVE LOGITS

 lure

1.51

 lured

1.30

GGGGGGGG

1.01

 prey

0.85

 bait

0.82

 irresist

0.80

wcs

0.77

 tempt

0.77

 enticing

0.77

glers

0.77

Activations Density 0.009%

words associated with attracting or persuading someone to do something

terms related to attracting or drawing in individuals

No Comments

No Known Activations