INDEX

Explanations

references to representation or acting on someone's behalf

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

ULTY

-0.54

 easily

-0.48

Ein

-0.47

-0.46

روز

-0.45

no

-0.44

Gus

-0.44

amak

-0.44

far

-0.43

cru

-0.43

POSITIVE LOGITS

 Representing

1.51

 mewakili

1.46

Represent

1.42

 representing

1.42

 Represent

1.40

 représentants

1.39

representing

1.37

represent

1.36

representative

1.36

 representative

1.35

Activations Density 0.178%

references to representation or acting on someone's behalf

No Comments

No Known Activations