INDEX

Explanations

mentions of advanced planning or warning

oai_token-act-pair · gpt-3.5-turbo

instances of the word "advance"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ters

-0.79

tics

-0.70

chens

-0.68

kered

-0.67

ãĤ´ãĥ³

-0.66

ars

-0.64

arus

-0.64

============

-0.64

ById

-0.64

ter

-0.63

POSITIVE LOGITS

 directives

0.84

ments

0.84

heit

0.80

 knowledge

0.75

 toward

0.74

chnology

0.73

 technology

0.72

 towards

0.71

uve

0.70

ment

0.70

Activations Density 0.049%

mentions of advanced planning or warning

instances of the word "advance"

No Comments

No Known Activations