INDEX

Explanations

references to individuals and their actions in a narrative context

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

PropertyChanging

-0.53

########.

-0.50

AttributeSet

-0.48

makeText

-0.47

SpringRunner

-0.44

sizeCache

-0.44

neſs

-0.44

AppMethodBeat

-0.44

WireFormatLite

-0.43

uros

-0.42

POSITIVE LOGITS

 дописавши

0.44

Emin

0.43

spot

0.38

lobe

0.37

Rhestr

0.37

évaluateur

0.37

contents

0.36

 للمعارف

0.36

cèse

0.35

 puissiez

0.34

Activations Density 0.374%

references to individuals and their actions in a narrative context

No Comments

No Known Activations