INDEX

Explanations

constructs related to programming data structures or elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

béco

-0.41

ervare

-0.39

nestjs

-0.38

SharedDtor

-0.38

inguém

-0.36

RAINT

-0.35

︿

-0.35

TintMode

-0.34

Iy

-0.34

oves

-0.34

POSITIVE LOGITS

 Paglinawan

0.45

erevan

0.44

.*")]

0.43

ByVersion

0.42

Viited

0.42

₡

0.42

AndEndTag

0.42

ftagPool

0.41

ModelMap

0.41

matchCondition

0.41

Activations Density 0.050%

constructs related to programming data structures or elements

No Comments

No Known Activations