INDEX

Explanations

instances of the word "when" used to indicate temporal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

 bezeichneter

-0.76

Fid

-0.68

 Indah

-0.67

Hush

-0.66

曖昧さ回避

-0.66

 Спасибо

-0.64

KommentareTeilen

-0.63

 Shrewsbury

-0.62

افظة

-0.62

 bezit

-0.62

POSITIVE LOGITS

 when

1.44

 WHEN

1.39

when

1.39

 When

1.34

When

1.29

WHEN

1.29

cuando

1.13

 när

1.11

Cuando

1.10

Khi

1.10

Activations Density 0.108%

instances of the word "when" used to indicate temporal context

No Comments

No Known Activations