INDEX

Explanations

phrases or constructions indicating conditions or hypothetical situations

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

as X

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Heff

-0.39

-0.37

-0.36

StackTrace

-0.34

-0.33

-0.32

expiration

-0.31

POSITIVE LOGITS

AddTagHelper

0.87

 Normdatei

0.84

 فريبيس

0.82

<_>

0.80

NameInMap

0.80

 autorytatywna

0.78

awtextra

0.77

 surla

0.76

 nakalista

0.75

として

0.75

Activations Density 0.033%

phrases or constructions indicating conditions or hypothetical situations

as X

No Comments

No Known Activations