INDEX

Explanations

references to various types of studios and laboratories

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

 accept

-0.42

LabelTagHelper

-0.41

 transfieras

-0.41

olastic

-0.40

 članak

-0.40

VERTIS

-0.40

 Accept

-0.38

 Geld

-0.38

bibfield

-0.38

Intervals

-0.38

POSITIVE LOGITS

parsedMessage

0.84

 studios

0.63

 studio

0.58

ImageContext

0.50

 kitchens

0.50

yntaxException

0.49

studios

0.48

 laboratories

0.46

 CanadaChoose

0.45

 laboratorium

0.45

Activations Density 0.016%

references to various types of studios and laboratories

No Comments

No Known Activations