INDEX

Explanations

greetings or terms of endearment used in correspondence

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ResponseWriter

-0.75

обходи

-0.70

 Seligman

-0.66

writeHead

-0.64

Hell

-0.64

hept

-0.63

 Dryden

-0.63

neider

-0.62

umbledore

-0.62

 Reverso

-0.62

POSITIVE LOGITS

 Dear

2.04

 dear

2.03

Dear

1.83

 DEAR

1.82

dear

1.81

DEAR

1.74

 dearest

1.22

Dearest

1.20

 beloved

0.95

亲爱的

0.91

Activations Density 0.112%

greetings or terms of endearment used in correspondence

No Comments

No Known Activations