INDEX

Explanations

markup and programming structure elements in code and document files

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 ModelExpression

-1.15

 CreateTagHelper

-1.09

 nahilalakip

-1.03

parsedMessage

-1.02

+#+#

-0.98

 transfieras

-0.95

featureID

-0.94

+#+

-0.93

 <<<<<<<<<<<<<<

-0.93

Datuak

-0.89

POSITIVE LOGITS

PR

0.37

AY

0.36

AL

0.36

auto

0.35

IM

0.34

AN

0.34

JE

0.34

OT

0.34

AM

0.34

Activations Density 0.280%

markup and programming structure elements in code and document files

No Comments

No Known Activations