INDEX

Explanations

references to needs and deficiencies in various contexts

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

mstyle

-0.60

 AppCompat

-0.52

AndEndTag

-0.50

 fingers

-0.49

SPARENT

-0.49

relsen

-0.46

 integrity

-0.46

 podjela

-0.46

 transpar

-0.46

 lucky

-0.46

POSITIVE LOGITS

 needs

1.27

Needs

1.21

 Needs

1.19

 demand

1.12

needs

1.12

 Bedarf

1.09

 need

1.08

 NEEDS

1.08

 besoins

1.08

需求

1.05

Activations Density 0.253%

references to needs and deficiencies in various contexts

No Comments

No Known Activations