INDEX

Explanations

phrases that indicate potential outcomes or possibilities in various contexts

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

preceding a verb or question

np_acts-logits-general · gemini-2.0-flash

seeks, obtained, closest, following, prob, value, described

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

Personendaten

-1.43

 kaarangay

-1.34

OGND

-1.26

 مشين

-1.25

 EconPapers

-1.25

featureID

-1.23

 AssemblyCulture

-1.23

rungsseite

-1.22

 مرئيه

-1.18

 tartalomajánló

-1.17

POSITIVE LOGITS

 again

0.99

0.83

↵

0.82

 Again

0.80

<strong>

0.78

0.77

 also

0.75

 επίσης

0.75

0.74

in

0.74

Activations Density 0.852%

phrases that indicate potential outcomes or possibilities in various contexts

preceding a verb or question

seeks, obtained, closest, following, prob, value, described

No Comments

No Known Activations