INDEX

Explanations

HTML header tags

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

HTML head and body

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

 queſta

-0.89

évaluateur

-0.77

$_(

-0.77

AndEndTag

-0.75

 Италијани

-0.73

 ſei

-0.73

 ſeine

-0.71

elemField

-0.71

 increí

-0.70

 beſte

-0.70

POSITIVE LOGITS

 initComponents

0.50

DockStyle

0.36

 BoxDecoration

0.35

</thead>

0.35

head

0.35

 head

0.33

ViewFeatures

0.33

setVerticalGroup

0.33

BeginInit

0.33

DOCTYPE

0.31

Activations Density 0.001%

HTML header tags

HTML head and body

No Comments

No Known Activations