INDEX

Explanations

references to past experiences and memories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/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.15.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

hasMoreElements

-0.52

 kasarigan

-0.47

WireFormatLite

-0.47

internalType

-0.47

ɵɵ

-0.45

 Majefty

-0.44

 Jefus

-0.44

UDC

-0.42

 onFailure

-0.42

 mukaan

-0.41

POSITIVE LOGITS

 yore

0.86

исленность

0.79

 تضيفلها

0.77

 früher

0.75

 vroeger

0.74

ArgsConstructor

0.70

 olden

0.70

AxisAlignment

0.68

 yester

0.68

 heyday

0.66

Activations Density 0.168%

references to past experiences and memories

No Comments

No Known Activations