INDEX

Explanations

references to mathematical concepts and definitions

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

Follows the word "The"

np_acts-logits-general · gemini-2.0-flash

the + specific noun

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

phosa

-0.53

 feeders

-0.52

[#

-0.52

 podr

-0.50

 Défense

-0.49

once

-0.49

 fér

-0.48

*[

-0.46

 rád

-0.46

 nadar

-0.45

POSITIVE LOGITS

 relation

0.91

SharedDtor

0.80

 total

0.77

 AssemblyCulture

0.75

 corresponding

0.73

 observed

0.73

 behavior

0.72

 resulting

0.69

енча

0.68

 range

0.68

Activations Density 1.645%

references to mathematical concepts and definitions

Follows the word "The"

the + specific noun

No Comments

No Known Activations