INDEX

Explanations

phrases emphasizing a specific location or point in a discussion

oai_token-act-pair · gpt-3.5-turbo

instances of the word "here," indicating a focus on emphasizing points or clarifying details within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 parap

-0.66

ews

-0.64

 Gujar

-0.60

 Mehran

-0.59

 Doors

-0.59

visors

-0.58

ggle

-0.58

 tongues

-0.56

uously

-0.55

amaz

-0.54

POSITIVE LOGITS

abouts

1.54

tics

1.54

tical

1.54

tic

1.20

upon

0.81

with

0.80

 guiActiveUn

0.78

after

0.73

from

0.70

here

0.69

Activations Density 0.061%

phrases emphasizing a specific location or point in a discussion

instances of the word "here," indicating a focus on emphasizing points or clarifying details within a context

No Comments

No Known Activations