INDEX

Explanations

references to political topics and discussions

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

time and date stamps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

RegressionTest

-0.73

 betweenstory

-0.68

%"),

-0.66

esModule

-0.63

nthesis

-0.61

frastructure

-0.61

::$_

-0.58

Efq

-0.58

JpaRepository

-0.58

recated

-0.58

POSITIVE LOGITS

IMPORTED

0.52

utilisons

0.52

enumii

0.51

 ujednoznacz

0.49

enumi

0.46

 MaterialApp

0.45

новниш

0.45

 acepción

0.44

ỡng

0.43

 calibre

0.41

Activations Density 0.015%

references to political topics and discussions

time and date stamps

No Comments

No Known Activations