INDEX

Explanations

references to spokespersons and officials

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

-0.62

-0.55

-0.53

an

-0.46

 alap

-0.46

 حلم

-0.44

To

-0.43

 Weir

-0.43

-0.42

POSITIVE LOGITS

 spokesman

2.62

 Spokes

2.49

 spokesperson

2.44

 spokeswoman

2.35

 spokespersons

2.13

 spokes

2.04

 portavoz

1.63

Spo

1.39

Efq

1.23

 themſelves

1.09

Activations Density 0.092%

references to spokespersons and officials

No Comments

No Known Activations