INDEX

Explanations

predicates indicating a temporal or numerical range or limit

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 extranjera

-0.48

 świą

-0.46

 asesinado

-0.44

 aihe

-0.44

 laikā

-0.42

 vandens

-0.41

 fieltro

-0.41

 mentre

-0.40

 olvido

-0.40

 która

-0.40

POSITIVE LOGITS

upto

0.68

 upto

0.67

 Unto

0.64

Hasta

0.60

 Hasta

0.59

 Leeds

0.59

Up

0.57

Up

0.57

########.

0.57

 Maxi

0.56

Activations Density 0.012%

predicates indicating a temporal or numerical range or limit

No Comments

No Known Activations