INDEX

Explanations

punctuation and formatting characters that separate or organize information

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

 myſelf

-0.71

 bershka

-0.71

 purpoſe

-0.67

 ſhe

-0.66

 ſeveral

-0.66

 Anſ

-0.65

 itſelf

-0.65

 Perſ

-0.64

 stanov

-0.64

ſy

-0.63

POSITIVE LOGITS

ValueStyle

0.87

0.60

__':

0.59

\":

0.59

TabIndex

0.54

 CreateTagHelper

0.54

SourceChecksum

0.53

@"/

0.53

abestanden

0.52

брь

0.51

Activations Density 0.407%

punctuation and formatting characters that separate or organize information

No Comments

No Known Activations