INDEX

Explanations

references to programming code and related structures, particularly focusing on variable and function names

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

']>;

-0.50

kháu

-0.45

UKAN

-0.44

[])

-0.44

 tradition

-0.43

 Qaraldi

-0.43

']))

-0.43

apun

-0.42

ongan

-0.42

faz

-0.42

POSITIVE LOGITS

verwijspagina

1.18

 للمعارف

0.94

ագրություններ

0.81

 AssemblyCulture

0.79

 Baillargeon

0.79

Efq

0.76

ArgsConstructor

0.75

SequentialGroup

0.72

 ujednoznacz

0.70

 IndexPath

0.67

Activations Density 0.201%

references to programming code and related structures, particularly focusing on variable and function names

No Comments

No Known Activations