INDEX

Explanations

any words that include the sequence "iti" followed by a high activation value of 7 or more

oai_token-act-pair · gpt-3.5-turbo

terms related to a specific organization or place called "Citi"

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

dress

-0.82

layer

-0.76

stroke

-0.74

worthy

-0.74

stairs

-0.73

meyer

-0.70

lessness

-0.64

fully

-0.63

sworth

-0.63

scale

-0.62

POSITIVE LOGITS

Äĩ

1.18

atives

1.05

iti

0.99

Ã±a

0.96

 Bike

0.88

emi

0.84

zed

0.83

Ã±

0.83

uting

0.83

ative

0.83

Activations Density 0.015%

any words that include the sequence "iti" followed by a high activation value of 7 or more

terms related to a specific organization or place called "Citi"

No Comments

No Known Activations