INDEX

Explanations

instances of examples and features in the text

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

:✨

-0.53

LookAnd

-0.33

 Normdatei

-0.33

anneer

-0.32

 bands

-0.31

getMenuInflater

-0.31

 لينك

-0.30

tagHelperRunner

-0.30

 mines

-0.29

 jura

-0.28

POSITIVE LOGITS

httphttps

0.62

omeness

0.54

#+#

0.54

Erreferentziak

0.52

 مثال

0.50

avía

0.49

 exemple

0.48

StructEnd

0.47

tangentMode

0.47

 instance

0.45

Activations Density 0.082%

instances of examples and features in the text

No Comments

No Known Activations