INDEX

Explanations

the particle "で," indicating locations or means in sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

"];

-0.70

 يتيمه

-0.68

 Holtz

-0.64

什么呢

-0.62

見えない

-0.62

見えます

-0.61

Datuak

-0.60

 iſt

-0.60

はこんな感じ

-0.57

ſhip

-0.57

POSITIVE LOGITS

で

3.51

で

2.36

でお

1.73

で大

1.71

りで

1.68

でも

1.63

では

1.62

などで

1.61

にて

1.53

での

1.50

Activations Density 0.015%

the particle "で," indicating locations or means in sentences

No Comments

No Known Activations