INDEX

Explanations

identifiers or codes related to data entries

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

elfde

-0.49

WebVitals

-0.49

θε

-0.49

setViewportView

-0.48

umina

-0.48

SourceChecksum

-0.48

aguas

-0.47

NSYLVANIA

-0.47

UDES

-0.47

 باسم

-0.46

POSITIVE LOGITS

Билгалдахарш

0.69

Архівовано

0.66

 protoimpl

0.62

oneofs

0.59

esgue

0.59

 Dunes

0.57

AccessorTable

0.56

ніципалі

0.56

antMatchers

0.55

DebuggerNonUser

0.54

Activations Density 0.002%

identifiers or codes related to data entries

No Comments

No Known Activations