INDEX

Explanations

technical terms and instructions related to software usage and settings

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

Followed by "and" or "under"

np_acts-logits-general · gemini-2.0-flash

Click, select, under, find

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ItemBackground

-0.69

AutoresizingMask

-0.66

 насељу

-0.63

 linkovi

-0.62

TagMode

-0.62

angliski

-0.62

tagHelperRunner

-0.59

inguém

-0.58

encodeWith

-0.57

ContentAsync

-0.57

POSITIVE LOGITS

 under

1.36

under

1.19

Under

1.08

 Under

1.03

 scroll

0.93

 UNDER

0.92

 unter

0.89

 select

0.89

 bajo

0.84

 click

0.83

Activations Density 0.183%

technical terms and instructions related to software usage and settings

Followed by "and" or "under"

Click, select, under, find

No Comments

No Known Activations