INDEX

Explanations

phrases related to physical actions involving objects or body parts

oai_token-act-pair · gpt-3.5-turbo

the conjunction "and" used in various contexts

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

ajor

-0.75

onymous

-0.73

actionDate

-0.72

uit

-0.72

urd

-0.71

uty

-0.69

Times

-0.69

busters

-0.67

aretz

-0.67

reality

-0.66

POSITIVE LOGITS

 then

1.11

 THEN

1.02

 secondly

0.98

 thence

0.95

 consequently

0.94

 thus

0.93

 proceeded

0.90

 thereby

0.88

 therefore

0.86

 gave

0.85

Activations Density 0.490%

phrases related to physical actions involving objects or body parts

the conjunction "and" used in various contexts

No Comments

No Known Activations