INDEX

Explanations

references to a specific individual or product

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

wad, VAD, evade

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 kasarigan

-0.87

styleType

-0.74

PreferredItem

-0.67

 '\\;'

-0.62

 otomatig

-0.61

createConnection

-0.60

-0.59

 препратки

-0.58

yyj

-0.56

 nakalista

-0.53

POSITIVE LOGITS

Vad

1.57

Wad

1.52

Vad

1.47

vad

1.47

vad

1.46

Wad

1.35

wad

1.30

wad

1.27

vada

0.96

vat

0.82

Activations Density 0.008%

references to a specific individual or product

wad, VAD, evade

No Comments

No Known Activations