INDEX

Explanations

references to testing and analysis

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

investigation and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

findpost

-0.45

RegressionTest

-0.40

 Administrativna

-0.38

hörigen

-0.33

upné

-0.32

تقاوى

-0.31

hoeddwyd

-0.31

ade

-0.31

nestjs

-0.30

ơi

-0.30

POSITIVE LOGITS

SharedCtor

0.58

 Normdatei

0.58

Hentet

0.57

parsedMessage

0.57

TagMode

0.57

 fubject

0.54

RTHOOK

0.50

 TextAppearance

0.50

 imagui

0.50

 Uploaded

0.50

Activations Density 0.000%

references to testing and analysis

investigation and analysis

No Comments

No Known Activations