INDEX

Explanations

phrases indicating uniqueness, rarity, or exceptional qualities

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

 himo

-0.74

PerformLayout

-0.69

Aholisi

-0.64

 >=",

-0.63

 CURIAM

-0.60

}")

-0.60

AutoScaleMode

-0.59

TemporalType

-0.57

 allemaal

-0.56

 alternates

-0.56

POSITIVE LOGITS

 unique

1.33

unique

1.09

 Unique

1.04

Unique

1.03

 rare

1.01

 extraordinary

0.99

 unprecedented

0.99

 unusual

0.99

 uniqueness

0.98

 уника

0.98

Activations Density 0.350%

phrases indicating uniqueness, rarity, or exceptional qualities

No Comments

No Known Activations