INDEX

Explanations

words related to specific locations, particularly ones containing the letter "w"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "w"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 mosqu

-0.87

 distingu

-0.78

milo

-0.75

 sensitivity

-0.73

 paraly

-0.70

 clen

-0.70

 ancest

-0.70

 susceptibility

-0.69

 fracturing

-0.68

 hemor

-0.68

POSITIVE LOGITS

itness

1.38

atts

1.38

elcome

1.26

ashington

1.21

izard

1.19

restling

1.16

edge

1.12

ITNESS

1.10

irts

1.10

arp

1.10

Activations Density 0.026%

words related to specific locations, particularly ones containing the letter "w"

occurrences of the letter "w"

No Comments

No Known Activations