INDEX

Explanations

instances of the word "original" and related forms, indicating a focus on foundational documents or concepts

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

original followed by noun

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

RenderAtEndOf

-0.53

TagMode

-0.50

zzleHttp

-0.49

UnusedPrivate

-0.44

 resourceCulture

-0.43

DrawerToggle

-0.40

DateTimeField

-0.40

setVerticalGroup

-0.39

<?=

-0.39

 latest

-0.39

POSITIVE LOGITS

 original

0.99

 originally

0.96

Original

0.90

 ORIGINAL

0.89

 Original

0.89

original

0.88

 Originally

0.88

 ursprünglichen

0.87

originally

0.86

Originally

0.83

Activations Density 0.052%

instances of the word "original" and related forms, indicating a focus on foundational documents or concepts

original followed by noun

No Comments

No Known Activations