INDEX

Explanations

phrases indicating specific occurrences or referents in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 more

-0.54

tamment

-0.53

 antaranya

-0.50

 sonst

-0.50

 ninh

-0.49

ampingi

-0.49

êmio

-0.48

sonian

-0.48

Contactez

-0.47

 rarement

-0.46

POSITIVE LOGITS

この

0.70

這一

0.69

 este

0.68

ఈ

0.67

 kasarigan

0.67

ఈ

0.67

 この

0.66

ഈ

0.66

Этот

0.65

 kind

0.65

Activations Density 0.137%

phrases indicating specific occurrences or referents in context

No Comments

No Known Activations