INDEX

Explanations

phrases or sentences where something is being named or labeled

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "calling" or labeling something

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

edia

-0.82

ockets

-0.74

bilt

-0.72

abal

-0.69

EEE

-0.66

taboola

-0.66

etheus

-0.65

ersen

-0.62

idth

-0.61

emen

-0.61

POSITIVE LOGITS

 bluff

0.85

"#

0.73

selves

0.69

ãĥ¼ãĥ³

0.66

``

0.66

ãĥ¼ãĥ

0.65

''

0.61

Activations Density 0.082%

phrases or sentences where something is being named or labeled

references to the concept of "calling" or labeling something

No Comments

No Known Activations