INDEX

Explanations

phrases indicating certainty or importance

oai_token-act-pair · gpt-3.5-turbo

the phrase "cannot be," indicating a focus on limitations or denials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Notting

-0.67

phabet

-0.64

endiary

-0.63

ipop

-0.62

inally

-0.59

 Patrol

-0.58

hyde

-0.58

pour

-0.56

culosis

-0.56

CFR

-0.56

POSITIVE LOGITS

 bothered

0.87

 forgiven

0.87

fits

0.84

 blamed

0.82

 undone

0.82

 construed

0.81

amed

0.78

 seen

0.77

witch

0.77

aming

0.76

Activations Density 0.065%

phrases indicating certainty or importance

the phrase "cannot be," indicating a focus on limitations or denials

No Comments

No Known Activations