INDEX

Explanations

the term "ents" indicating a focus on entity descriptors or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 snippetHide

-0.55

XmlAccessType

-0.47

 księ

-0.46

 packaged

-0.46

Bronnen

-0.45

 дей

-0.44

Front

-0.43

eqref

-0.43

Mun

-0.42

ction

-0.42

POSITIVE LOGITS

 swears

0.99

 swear

0.95

 CreateTagHelper

0.94

 swearing

0.85

 ModelExpression

0.81

 vows

0.77

 oath

0.75

makeText

0.73

 sworn

0.73

 swore

0.72

Activations Density 0.058%

the term "ents" indicating a focus on entity descriptors or classifications

No Comments

No Known Activations