INDEX

Explanations

words related to a negative reaction or complaining

oai_token-act-pair · gpt-3.5-turbo

references to excessive commotion or concern

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

vae

-0.71

 prisoner

-0.66

 prison

-0.66

ACTED

-0.64

ramer

-0.64

rase

-0.62

ramid

-0.62

 prisoners

-0.60

rams

-0.60

 Prison

-0.59

POSITIVE LOGITS

naire

1.06

iness

1.00

 fuss

0.95

naires

0.94

engers

0.94

ãĤ¦ãĤ¹

0.89

liness

0.85

eful

0.83

ing

0.82

 Leilan

0.82

Activations Density 0.012%

words related to a negative reaction or complaining

references to excessive commotion or concern

No Comments

No Known Activations