INDEX

Explanations

lines or sections of documentation marked by asterisks or similar visual separators

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

user

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

 inne

-0.40

י

-0.34

 Sinf

-0.33

 velkommen

-0.33

vignon

-0.33

 incor

-0.33

 incomes

-0.32

 canto

-0.32

раи

-0.32

含

-0.31

POSITIVE LOGITS

*********

1.65

********

1.06

ScopeManager

0.86

httphttps

0.85

**********

0.81

StructEnd

0.78

 ********

0.77

*************

0.77

uxxxx

0.76

************

0.75

Activations Density 0.000%

lines or sections of documentation marked by asterisks or similar visual separators

user

No Comments

No Known Activations