INDEX

Explanations

words related to professions and work environments

oai_token-act-pair · gpt-3.5-turbo

phrases related to law and authority figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 lamented

-0.61

 Â®

-0.60

CLS

-0.60

arnaev

-0.59

 eagerly

-0.59

 mere

-0.58

 famed

-0.58

 Released

-0.58

"#

-0.57

surprisingly

-0.57

POSITIVE LOGITS

1.26

['

1.25

 everybody

1.10

 âĢ¦"

1.08

 ..."

1.07

,"

1.07

 somebody

1.06

.''

1.06

."

1.05

,'"

1.05

Activations Density 1.337%

words related to professions and work environments

phrases related to law and authority figures

No Comments

No Known Activations