INDEX

Explanations

references to specific classifications or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

adone

-0.74

ródło

-0.71

 işe

-0.68

umenical

-0.68

tigas

-0.66

 Brownian

-0.65

 bombe

-0.65

 solusi

-0.65

 pracy

-0.65

 Fläche

-0.61

POSITIVE LOGITS

 certain

1.84

 Certain

1.82

Certain

1.71

certain

1.64

 CERTAIN

1.58

 ciertos

1.47

 ciertas

1.36

 certaine

1.36

 bepaalde

1.19

 certa

1.14

Activations Density 0.080%

references to specific classifications or categories

No Comments

No Known Activations