INDEX

Explanations

references to broadness or inclusivity in terms of scope 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

Utf

-0.84

Jai

-0.78

Ghi

-0.76

 Pelham

-0.75

 ronde

-0.74

Jai

-0.74

 cages

-0.74

 chivalry

-0.73

ahi

-0.73

 Eskimo

-0.72

POSITIVE LOGITS

 Broad

1.59

Broad

1.52

broad

1.44

 broad

1.29

 BROAD

1.24

 broader

1.17

 broadly

1.09

BROAD

1.05

 broadened

1.03

 Brod

0.95

Activations Density 0.125%

references to broadness or inclusivity in terms of scope or categories

No Comments

No Known Activations