INDEX

Explanations

references to software versions and updates

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

ſelf

-0.75

 myſelf

-0.73

 للاسماء

-0.71

 esternos

-0.68

 ſche

-0.66

ChildScrollView

-0.65

 '\\;'

-0.65

Diweddarwch

-0.64

 pleaſure

-0.64



-0.64

POSITIVE LOGITS

 version

0.30

 Gottlieb

0.29

kot

0.29

°

0.29

 жар

0.28

 ánh

0.28

 Hickman

0.28

kel

0.28

 recenti

0.28

前

0.28

Activations Density 0.041%

references to software versions and updates

No Comments

No Known Activations