INDEX

Explanations

references to data structures and functions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

:✨

-0.56

 resourceCulture

-0.54

webElementXpaths

-0.52

 Pendleton

-0.51

msgTypes

-0.50

 Hodg

-0.49

 Weston

-0.49

ValueGeneration

-0.49

 Hodgson

-0.49

 <>",

-0.49

POSITIVE LOGITS

 Jarvis

0.92

zar

0.91

lar

0.90

JAR

0.90

zar

0.89

JAR

0.89

Yar

0.88

LAR

0.88

jar

0.88

SAR

0.86

Activations Density 0.691%

references to data structures and functions

No Comments

No Known Activations