INDEX

Explanations

punctuation marks, particularly periods and commas

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

Text after colons or dashes

np_acts-logits-general · gemini-2.0-flash

introductions and summaries

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 firstly

-0.80

 entweder

-0.77

 either

-0.77

 Firstly

-0.74

一是

-0.74

 nämlich

-0.72

either

-0.71

 Specifically

-0.69

 particularly

-0.69

 specifically

-0.68

POSITIVE LOGITS

Bref

1.19

etc

1.16

etc

1.15

Etc

1.13

Etc

1.08

 tudo

1.06

 словом

1.06

 bref

1.05

总之

1.03

这一切

1.02

Activations Density 0.254%

punctuation marks, particularly periods and commas

Text after colons or dashes

introductions and summaries

No Comments

No Known Activations