INDEX

Explanations

references to historical texts and authorship

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

before a publication type

np_acts-logits-general · gemini-2.0-flash

written or published works

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

zzleHttp

-0.64

GTCX

-0.57

ConstraintMaker

-0.48

 intptr

-0.48

InitStruct

-0.47

Autoritní

-0.47

 disambiguazione

-0.47

Cyfeiriadau

-0.47

 CreateTagHelper

-0.46

$_['

-0.45

POSITIVE LOGITS

 written

0.49

 published

0.47

 rédig

0.45

written

0.44

 publication

0.44

 series

0.44

 fonti

0.43



0.42

published

0.42

 gepubliceerd

0.42

Activations Density 0.272%

references to historical texts and authorship

before a publication type

written or published works

No Comments

No Known Activations