INDEX

Explanations

named entities, particularly places and official titles

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

SLP

-0.61

__':

-0.60

Curi

-0.56

#![

-0.53

ixos

-0.51

 Lula

-0.51

 BoxFit

-0.51

uskas

-0.51

SLP

-0.50

GNUC

-0.50

POSITIVE LOGITS

mandu

1.16

 Nepal

1.05

Nepal

1.04

 Kathmandu

1.03

 Nepali

0.92

 Vermont

0.77

Vermont

0.72

XtraGrid

0.67

Rwanda

0.66

 Adirond

0.64

Activations Density 0.523%

named entities, particularly places and official titles

No Comments

No Known Activations