INDEX

Explanations

programming or code-related structures and syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

writeFieldEnd

-0.68

CppCodeGen

-0.68

NamedQueries

-0.62

 beginnetje

-0.59

Попис

-0.57

ällor

-0.56

 xlink

-0.55

GEBURTSDATUM

-0.55

setObjectName

-0.54

astéroïdes

-0.53

POSITIVE LOGITS

ArgsConstructor

0.66

BeginContext

0.57

 للمعارف

0.57

erunner

0.55

JAMIN

0.55

 EdgeInsets

0.54

Билгалдахарш

0.54

 kneeling

0.51

 Mounts

0.51

 Rien

0.50

Activations Density 0.021%

programming or code-related structures and syntax

No Comments

No Known Activations