INDEX

Explanations

sections of the document that are formatted as code or technical instructions

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

Word fragments or beginning of words

np_acts-logits-general · gemini-2.0-flash

marine words

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

-0.75

“

-0.73

„

-0.67

!”

-0.66

‘

-0.64

-0.63

-0.61

!"

-0.57

“

-0.56

!“

-0.54

POSITIVE LOGITS

parsedMessage

1.12

脚注の使い方

1.10

IsMutable

1.08

 محفوظة

1.07

 Мексичка

0.98

 مرئيه

0.95

 CreateTagHelper

0.94

Personensuche

0.91

Portale

0.91

 समीक्षाओं

0.89

Activations Density 0.038%

sections of the document that are formatted as code or technical instructions

Word fragments or beginning of words

marine words

No Comments

No Known Activations