INDEX

Explanations

phrases that indicate a recent change or addition, particularly with the word "newly."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 vectorielles

-0.82

__((

-0.71

Dislikes

-0.65

 Méndez

-0.63

milk

-0.63

 lumineuse

-0.62

orkin

-0.61

 filtre

-0.60

 Mondial

-0.59

olulu

-0.59

POSITIVE LOGITS

stateProvider

0.68

 lenker

0.63

ThroughAttribute

0.60

styleType

0.60

 <=",

0.58

)':

0.57

DebuggerNonUser

0.57

 thrill

0.53

以

0.52

astify

0.52

Activations Density 0.168%

phrases that indicate a recent change or addition, particularly with the word "newly."

No Comments

No Known Activations