INDEX

Explanations

quantifiable data and information from studies or analyses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 над

-0.28

my

-0.28

if

-0.27

<eos>

-0.24

之

-0.24

once

-0.22

my

-0.21

 over

-0.21

POSITIVE LOGITS

AnchorStyles

0.93

 للاسماء

0.92

bootstrapcdn

0.90

msgTypes

0.85

QMetaType

0.84

MessageTagHelper

0.84

majánló

0.81

setVerticalGroup

0.81

<unused8>

0.80

<unused14>

0.80

Activations Density 0.969%

quantifiable data and information from studies or analyses

No Comments

No Known Activations