INDEX

Explanations

sentences related to procedural steps or guidelines

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

Followed by the word "Test" or "test"

np_acts-logits-general · gemini-2.0-flash

testTesttesting

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PropertyGroup

-0.61

autoconfigure

-0.61

DockStyle

-0.58

twimg

-0.55

colgroup

-0.54

cloudfront

-0.53

VIAF

-0.53

AndEndTag

-0.52

帖最后由

-0.51

ActivityCompat

-0.50

POSITIVE LOGITS

 test

2.80

 Test

2.58

Test

2.47

test

2.45

 testing

2.39

 TEST

2.38

 tests

2.37

TEST

2.21

 Tests

2.19

 Testing

2.19

Activations Density 0.143%

sentences related to procedural steps or guidelines

Followed by the word "Test" or "test"

testTesttesting

No Comments

No Known Activations