INDEX

Explanations

phrases that indicate causal relationships or dependencies between variables

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

cause and effect relationships

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

geslacht

-0.44

$('.

-0.37

 prom

-0.35

 Sodom

-0.34

 chó

-0.33

 البريط

-0.33

 sections

-0.33

 gerecht

-0.33

性

-0.32

bli

-0.32

POSITIVE LOGITS

 apparently

0.63

apparently

0.58

EDEFAULT

0.56

windowFixed

0.54

 offenbar

0.54

 intptr

0.54

localctx

0.52

verifyException

0.52

 HasFactory

0.52

DockStyle

0.52

Activations Density 0.153%

phrases that indicate causal relationships or dependencies between variables

cause and effect relationships

No Comments

No Known Activations