INDEX

Explanations

references to the concept of "front" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ukone

-0.60

 entendu

-0.56

 noires

-0.56

 sûr

-0.55

 okuyayım

-0.55

Tribe

-0.54

 électron

-0.54

 culturelles

-0.53

 foncé

-0.53

TintMode

-0.53

POSITIVE LOGITS

 front

0.83

 FRONT

0.81

 Front

0.81

Front

0.73

ंदीखरीदारी

0.71

front

0.70

FRONT

0.68

ViewFeatures

0.67

 GetEnumerator

0.66

Fron

0.65

Activations Density 0.055%

references to the concept of "front" in various contexts

No Comments

No Known Activations