INDEX

Explanations

instances of formal tokens or tags often used in programming or markup languages

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

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

modb

-0.69

BeginContext

-0.65

раздо

-0.60

ArrowToggle

-0.59

InstanceState

-0.56

WithEmail

-0.55

 становника

-0.55

 виправивши

-0.55

<!--[

-0.54

HtmlAttribute

-0.54

POSITIVE LOGITS

 they

1.54

it

1.35

we

1.32

you

1.16

he

1.07

she

1.05

 они

0.90

the

0.81

 these

0.77

 она

0.77

Activations Density 0.114%

instances of formal tokens or tags often used in programming or markup languages

No Comments

No Known Activations