INDEX

Explanations

phrases indicating comparisons or contrasts

oai_token-act-pair · gpt-3.5-turbo

themes related to societal issues and power dynamics

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

keeper

-0.74

ottest

-0.71

ricanes

-0.70

ired

-0.70

scrib

-0.69

kies

-0.68

ilant

-0.67

sites

-0.67

Syrian

-0.66

allel

-0.66

POSITIVE LOGITS

 causation

1.30

 nurture

1.27

 spirituality

1.23

 emotion

1.23

 ethnicity

1.20

 charisma

1.18

 ideology

1.17

 sexuality

1.17

 destiny

1.16

 morality

1.16

Activations Density 0.308%

phrases indicating comparisons or contrasts

themes related to societal issues and power dynamics

No Known Activations