INDEX

Explanations

phrases related to cost or value, particularly when associated with goods or services

oai_token-act-pair · gpt-3.5-turbo

references to affordability or low cost

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

âĹ¼

-0.90

vironment

-0.83

ittee

-0.82

uberty

-0.77

ittees

-0.76

mbuds

-0.75

yss

-0.72

ADRA

-0.72

ational

-0.71

 skelet

-0.71

POSITIVE LOGITS

ened

1.08

liest

1.01

eners

0.92

 cheap

0.92

lier

0.89

ening

0.81

 enough

0.80

ener

0.79

priced

0.76

ness

0.75

Activations Density 0.013%

phrases related to cost or value, particularly when associated with goods or services

references to affordability or low cost

No Comments

No Known Activations