INDEX

Explanations

specific formatting and notation elements commonly used in mathematical expressions or programming syntax

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

Text after numbers, symbols, or uncommon characters

np_acts-logits-general · gemini-2.0-flash

curves

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

ad

-0.84

par

-0.76

in

-0.69

HasForeignKey

-0.69

qu

-0.67

-0.66

ruptedException

-0.66

op

-0.66

帖最后由

-0.66

re

-0.65

POSITIVE LOGITS

 purpoſe

1.67

 myſelf

1.65

 Monfieur

1.59

 Jefus

1.58

 reaſon

1.55

 pleaſure

1.54

 Theſe

1.53

 Anſ

1.52

 ſeveral

1.51

 ſtate

1.51

Activations Density 0.894%

specific formatting and notation elements commonly used in mathematical expressions or programming syntax

Text after numbers, symbols, or uncommon characters

curves

No Comments

No Known Activations