INDEX

Explanations

phrases related to statements or declarations made by individuals

oai_token-act-pair · gpt-3.5-turbo

the preposition "in" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ens

-0.72

wagon

-0.71

 bandwagon

-0.71

 learners

-0.65

 fared

-0.64

 looms

-0.64

 lodged

-0.63

 Mages

-0.62

killed

-0.62

 therein

-0.61

POSITIVE LOGITS

 lieu

1.11

ordinate

1.02

clus

1.01

 unison

1.00

clusively

0.99

humane

0.96

 order

0.95

 conjunction

0.94

ventory

0.93

clusions

0.92

Activations Density 0.163%

phrases related to statements or declarations made by individuals

the preposition "in" across various contexts

No Comments

No Known Activations