INDEX

Explanations

numerical values that likely represent contact details or identifiers

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 itſelf

-0.95

archiviato

-0.94

WebElementEntity

-0.91

rrggbb

-0.89

 '\\;'

-0.88

 photolibrary

-0.88

 myſelf

-0.88

tanleria

-0.88

webElementXpaths

-0.86

 tfsi

-0.85

POSITIVE LOGITS

0.63

0.53

↵↵

0.46

0.44

0.42

0.41

–

0.41

Activations Density 0.042%

numerical values that likely represent contact details or identifiers

No Comments

No Known Activations