INDEX

Explanations

instances of educational or explanatory content regarding various subjects

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

Follows a punctuation mark or article

np_acts-logits-general · gemini-2.0-flash

different resolutions or parts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

__':

-0.85

__':

-0.74

apparently

-0.66

Apparently

-0.65

 nominally

-0.64

 Apparently

-0.63

 apparently

-0.62

ècie

-0.62

toir

-0.59

řejmě

-0.59

POSITIVE LOGITS

 compito

0.64

 réguli

0.64

 dévelo

0.63

ⓧ

0.62

 avulla

0.61

GenerationType

0.60

#+#

0.60

 rapidez

0.59

 hâte

0.58

 définiti

0.58

Activations Density 0.272%

instances of educational or explanatory content regarding various subjects

Follows a punctuation mark or article

different resolutions or parts

No Comments

No Known Activations