INDEX

Explanations

textual placeholders or indicators for omitted content

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 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

tvguidetime

-0.52

 redes

-0.45

Ded

-0.43

xt

-0.42

bal

-0.40

}>{

-0.40

 odkazy

-0.39

Literatur

-0.39

tas

-0.39

ConstraintMaker

-0.39

POSITIVE LOGITS

▼

0.97

localctx

0.82

 utafitiHapana

0.79

ंदीखरीदारी

0.72

ыгана

0.72

XmlAccessType

0.71

 Meksiku

0.70

anair

0.69

tanleria

0.68

theit

0.68

Activations Density 0.252%

textual placeholders or indicators for omitted content

No Comments

No Known Activations