INDEX

Explanations

terms associated with display performance and electronic components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

principalColumn

-0.62

rrggbb

-0.57

 propOrder

-0.57

Kea

-0.53

 assholes

-0.53

HasForeignKey

-0.52

 Савезне

-0.51

\{\\

-0.51

blurRadius

-0.51

awtextra

-0.50

POSITIVE LOGITS

 oneself

0.70

be

0.64

 dedans

0.60

 utmost

0.57

 ossia

0.55

 namely

0.55

 place

0.55

 asemenea

0.54

 somebody

0.54

하다

0.53

Activations Density 0.194%

terms associated with display performance and electronic components

No Comments

No Known Activations