INDEX

Explanations

words related to different levels of hierarchy or importance, with a focus on the upper ranks such as executives or leaders

oai_token-act-pair · gpt-3.5-turbo

references to ecological or environmental concepts

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

 Desire

-0.63

 Attempt

-0.61

FTWARE

-0.60

 Commit

-0.59

 dear

-0.59

NOW

-0.59

 Dull

-0.59

 Kitchen

-0.58

MENTS

-0.58

 rack

-0.58

POSITIVE LOGITS

uador

1.24

stasy

1.23

utive

1.12

lectic

1.11

oin

1.05

ogl

1.05

rypt

1.03

urrent

1.03

entric

0.99

chio

0.97

Activations Density 0.022%

words related to different levels of hierarchy or importance, with a focus on the upper ranks such as executives or leaders

references to ecological or environmental concepts

No Comments

No Known Activations