INDEX

Explanations

phrases that indicate conditions or eligibility factors

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

relation specific words

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

 intend

-0.34

land

-0.33

 habr

-0.33

’

-0.32

 encuent

-0.31

Log

-0.31

-0.30

Url

-0.30

hi

-0.30

POSITIVE LOGITS

GEBURTSDATUM

0.98

RegressionTest

0.94

BufferException

0.93

webElementXpaths

0.89

OGND

0.84

AndEndTag

0.83

Tikang

0.82

دانشنامهٔ

0.81

 فريبيس

0.81

 InputDecoration

0.80

Activations Density 0.079%

phrases that indicate conditions or eligibility factors

relation specific words

No Comments

No Known Activations