INDEX

Explanations

the word "adjust" followed by a preposition

oai_token-act-pair · gpt-3.5-turbo

phrases indicating an action or function related to "adjusting."

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

Dash

-0.63

iasco

-0.61

inguished

-0.60

 mysteriously

-0.58

Released

-0.57

mask

-0.56

haw

-0.55

 riddled

-0.54

imposed

-0.53

 Springer

-0.53

POSITIVE LOGITS

 oneself

0.72

 parity

0.68

=-=-

0.67

ordinate

0.65

anz

0.65

 anything

0.65

asted

0.64

role

0.63

=-=-=-=-

0.62

 whichever

0.61

Activations Density 0.701%

the word "adjust" followed by a preposition

phrases indicating an action or function related to "adjusting."

No Comments

No Known Activations