INDEX

Explanations

phrases related to following or subsequent actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "followed" and its context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

voy

-0.68

uci

-0.68

cit

-0.68

ukong

-0.65

flo

-0.64

tu

-0.64

venge

-0.64

eno

-0.63

utor

-0.63

ãĥīãĥ©ãĤ´ãĥ³

-0.62

POSITIVE LOGITS

 suit

0.99

ĸļ

0.95

 closely

0.92

suit

0.79

 faithfully

0.75

ictions

0.72

:]

0.71

SHIP

0.69

 logically

0.68

 thereafter

0.66

Activations Density 0.033%

phrases related to following or subsequent actions

instances of the word "followed" and its context

No Comments

No Known Activations