INDEX

Explanations

less than signs and angle brackets, often associated with code or markup elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

featureID

-0.77

 nakalista

-0.77

 resourceCulture

-0.68

elemField

-0.63

setVerticalGroup

-0.63

:])

-0.63

errHandler

-0.61

__':

-0.59

WithIOException

-0.58

 OMITBAD

-0.58

POSITIVE LOGITS

0.83

 $("<

0.70

 الحره

0.69

〈

0.67

0.66

 EdgeInsets

0.64

 Thin

0.62

 $('<

0.61

 Davison

0.61

antuan

0.59

Activations Density 0.023%

less than signs and angle brackets, often associated with code or markup elements

No Comments

No Known Activations