INDEX

Explanations

mentions of objects or items based on their size

oai_token-act-pair · gpt-3.5-turbo

references to size or dimensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

endez

-0.83

rique

-0.81

rea

-0.81

WP

-0.78

veyard

-0.77

onso

-0.76

agher

-0.76

perty

-0.74

ruption

-0.74

bara

-0.73

POSITIVE LOGITS

 scale

0.87

 stature

0.82

 mismatch

0.81

 circumference

0.80

 guiActiveUnfocused

0.79

 proportions

0.75

 sizes

0.75

 size

0.75

 scales

0.73

 sized

0.71

Activations Density 0.022%

mentions of objects or items based on their size

references to size or dimensions

No Comments

No Known Activations