INDEX

Explanations

forms of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

-1.03

bootstrapcdn

-1.00

 nakalista

-0.98

OGND

-0.92

 Allegretto

-0.90

 Jefus

-0.89

 '\\;'

-0.87

 betweenstory

-0.86

 Theſe

-0.85

 Theodo

-0.85

POSITIVE LOGITS

1.00

am

0.85

the

0.77

Am

0.72

im

0.72

an

0.69

Im

0.69

spac

0.68

0.67

us

0.65

Activations Density 0.135%

forms of the verb "to be."

No Comments

No Known Activations