INDEX

Explanations

instances of the word 'everything'

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "everything" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

idon

-0.76

geons

-0.72

ename

-0.71

odan

-0.68

dor

-0.67

eme

-0.66

chieve

-0.65

een

-0.65

orst

-0.64

furt

-0.63

POSITIVE LOGITS

 else

1.70

 imaginable

1.22

 Else

1.21

Else

1.08

 except

1.00

 revolves

0.92

else

0.90

 pertaining

0.87

 happening

0.87

 conceivable

0.82

Activations Density 0.056%

instances of the word 'everything'

references to the concept of "everything" in various contexts

No Comments

No Known Activations