INDEX

Explanations

terms related to women’s health and reproductive issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 himself

-1.16

himself

-0.95

 Himself

-0.85

__':

-0.82

__":

-0.81

 boyhood

-0.79

父子

-0.72

 doInBackground

-0.70

 koji

-0.69

łbym

-0.66

POSITIVE LOGITS

 herself

1.69

herself

1.25

she

1.25

her

1.21

 hennes

1.02

 ihrem

1.01

 haar

0.99

 její

0.96

 그녀

0.95

她在

0.94

Activations Density 1.911%

terms related to women’s health and reproductive issues

No Comments

No Known Activations