INDEX

Explanations

variations of the word "for" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ftagPool

-0.70

onekana

-0.66

 препратки

-0.66

Зноскі

-0.61

nanoTime

-0.57

 gada

-0.55

<?=

-0.55

ⓧ

-0.54

theless

-0.54

 تعدى

-0.54

POSITIVE LOGITS

 تضيفلها

0.71

 للاسماء

0.65

StoryboardSegue

0.59

JoinColumn

0.57

ific

0.57

 FormBuilder

0.56

lorette

0.56

dymo

0.56

fractive

0.55

lagem

0.54

Activations Density 0.023%

variations of the word "for" in different contexts

No Comments

No Known Activations