INDEX

Explanations

phrases emphasizing certainty or likelihood about events or observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 waking

-0.60

 bothering

-0.48

 looming

-0.46

 Herman

-0.45

 woken

-0.43

 sticking

-0.43

 blooming

-0.43

 benzer

-0.43

 staring

-0.42

 shaking

-0.42

POSITIVE LOGITS

Kjelder

0.89

avía

0.88

 antaranya

0.85

 nawr

0.85

Бахар

0.82

 IBOutlet

0.81

 티즈

0.79

 deſt

0.79

 raiſ

0.77

jalá

0.77

Activations Density 0.715%

phrases emphasizing certainty or likelihood about events or observations

No Comments

No Known Activations

phrases emphasizing certainty or likelihood about events or observations

No Comments

No Known Activations