INDEX

Explanations

phrases indicating time and location in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

parsedMessage

-0.62

titleMargin

-0.61

MemoryWarning

-0.57

ResumeLayout

-0.56

MergeFrom

-0.54

ReusableCell

-0.53

AsUp

-0.51

GIVEREF

-0.50

chism

-0.49

WriteLiteral

-0.49

POSITIVE LOGITS

 location

2.63

 locations

2.40

 Location

2.33

location

2.19

Location

2.14

 Locations

2.10

 LOCATION

2.10

Locations

2.00

locations

1.95

LOCATION

1.93

Activations Density 0.503%

phrases indicating time and location in various contexts

No Comments

No Known Activations