INDEX

Explanations

phrases related to additional information or actions

oai_token-act-pair · gpt-3.5-turbo

the phrase "in addition" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

blank

-0.65

gaard

-0.65

Hide

-0.62

assed

-0.60

 enthus

-0.60

owe

-0.59

Euro

-0.59

ouses

-0.58

Deal

-0.56

core

-0.56

POSITIVE LOGITS

 interestingly

0.83

 moreover

0.81

 anecd

0.80

 besides

0.78

 although

0.75

 secondly

0.74

 however

0.73

 according

0.70

 insofar

0.69

hester

0.68

Activations Density 0.136%

phrases related to additional information or actions

the phrase "in addition" and its variations

No Comments

No Known Activations