INDEX

Explanations

significant numerical values or annotations in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

guy

-0.80

 guys

-0.79

 basically

-0.76

 stuff

-0.74

-0.73

 pretty

-0.72

-0.69

 Basically

-0.68

-0.66

get

-0.64

POSITIVE LOGITS

__":

0.94

المناصب

0.88

 sonno

0.88

 templado

0.85

TestingModule

0.85

ArrowToggle

0.85

 feroit

0.84

ItemBackground

0.84

 FetchType

0.83

 métallique

0.83

Activations Density 0.672%

significant numerical values or annotations in text

No Comments

No Known Activations