INDEX

Explanations

instances of conditional phrases or situations involving "when" and actions taken by "you."

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

 propOrder

-0.77

NameInMap

-0.69

 transférez

-0.64

HasAnnotation

-0.64

互联网档案馆

-0.59

attore

-0.58

lenmiş

-0.57

Hush

-0.56

 Ewig

-0.56

 AppDelegate

-0.56

POSITIVE LOGITS

 when

1.18

when

1.16

 WHEN

1.11

When

1.10

 When

1.06

Cuando

1.04

cuando

1.02

WHEN

1.01

 cuando

1.01

 Cuando

0.97

Activations Density 0.232%

instances of conditional phrases or situations involving "when" and actions taken by "you."

No Comments

No Known Activations