INDEX

Explanations

unique identifiers or codes associated with specific data structures or entities

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

достатки

-0.81

ſo

-0.79

 ostavi

-0.78

ſa

-0.78

BibitemShut

-0.77

 houſe

-0.74

 Мексичка

-0.72

éphane

-0.70

)».

-0.70

extAlignment

-0.67

POSITIVE LOGITS

0.73

0.61

’

0.56

ССР

0.56

的

0.54

#+#

0.54

′

0.53

opsida

0.52

 Dade

0.50

ngilizce

0.50

Activations Density 0.660%

unique identifiers or codes associated with specific data structures or entities

No Comments

No Known Activations