INDEX

Explanations

words related to being 'like-minded'

oai_token-act-pair · gpt-3.5-turbo

references to individuals or groups that share similar beliefs or perspectives

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

IVER

-0.75

pees

-0.75

ARDS

-0.72

è¦ļéĨĴ

-0.72

adium

-0.70

PsyNetMessage

-0.70

ILA

-0.68

IDS

-0.68

apon

-0.67

osponsors

-0.67

POSITIVE LOGITS

ness

1.55

edly

1.17

nesses

1.13

minded

1.02

ly

1.00

 zeal

0.84

 minded

0.80

lihood

0.76

ingly

0.75

 prag

0.74

Activations Density 0.042%

words related to being 'like-minded'

references to individuals or groups that share similar beliefs or perspectives

No Comments

No Known Activations