INDEX

Explanations

return statements in code and associated structural elements

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

 '\\;'

-1.09

OGND

-0.92

 propOrder

-0.88

sizeCache

-0.86

 protoimpl

-0.85

basicConfig

-0.84

 للاسماء

-0.84

numerusform

-0.84

IsContent

-0.83

StructEnd

-0.83

POSITIVE LOGITS

textTheme

0.44

WriteLiteral

0.42

Begin

0.41

zeki

0.41

等到

0.40

 markup

0.40

una

0.39

 चला

0.38

cula

0.38

ROW

0.38

Activations Density 0.010%

return statements in code and associated structural elements

No Comments

No Known Activations