INDEX

Explanations

geographical locations and references to places within specific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ModelExpression

-0.84

SequentialGroup

-0.65

StructEnd

-0.63

WriteBarrier

-0.62

matchCondition

-0.61

writeField

-0.60

 Maren

-0.58

Deh

-0.58

iress

-0.58

WidgetItem

-0.58

POSITIVE LOGITS

 Brisbane

0.68

 Queensland

0.65

IVEREF

0.58

 Gold

0.56

ycz

0.55

Queensland

0.54

 Fraser

0.53

QLD

0.53

Brisbane

0.53

 Ipswich

0.52

Activations Density 0.091%

geographical locations and references to places within specific contexts

No Comments

No Known Activations