INDEX

Explanations

phrases related to location or place

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 zufolge

-0.50

use

-0.46

key

-0.41

 szerint

-0.41

 suatu

-0.40

nä

-0.40

 istot

-0.39

dbh

-0.39

 интересу

-0.39

 period

-0.39

POSITIVE LOGITS

ArgsConstructor

0.76

Personendaten

0.74

onAttach

0.72

__(/*!

0.70

 kasarigan

0.69

FormTagHelper

0.66

VersionUID

0.65

enumi

0.64

Externé

0.64

 beginnings

0.63

Activations Density 0.169%

phrases related to location or place

No Comments

No Known Activations