INDEX

Explanations

instances of the word "discover" in various contexts

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

the word "discover" or variations of it

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

yip

-0.79

annis

-0.79

hovah

-0.77

ortunately

-0.76

perty

-0.75

cius

-0.74

cise

-0.72

cussion

-0.72

pill

-0.71

cision

-0.71

POSITIVE LOGITS

ibility

0.99

ãĤ©

0.90

Discover

0.88

ãĥķãĤ¡

0.85

çīĪ

0.85

iveness

0.85

tainment

0.85

ä½¿

0.81

 Wond

0.80

ãĤ¦ãĤ¹

0.79

Activations Density 0.009%

instances of the word "discover" in various contexts

the word "discover" or variations of it

No Comments

No Known Activations