INDEX

Explanations

phrases indicating purpose or function

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

principalTable

-0.48

 ujednoznacz

-0.44

 CommonModule

-0.43

AutoScaleMode

-0.43

AutoScale

-0.42

 متعلقه

-0.42

 écoulé

-0.40

delwed

-0.40

ếu

-0.39

bleau

-0.39

POSITIVE LOGITS

OGND

0.55

hyrchwyd

0.41

formik

0.39

ungsver

0.37

例句

0.37

WithFormat

0.34

CompilerServices

0.33

pushFollow

0.33

fraz

0.33

decorators

0.33

Activations Density 0.154%

phrases indicating purpose or function

No Comments

No Known Activations