INDEX

Explanations

phrases that express temporal or spatial relationships

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 during

-0.55

 with

-0.55

 before

-0.54

 against

-0.54

 while

-0.51

 when

-0.49

 about

-0.48

on

-0.48

 within

-0.46

in

-0.44

POSITIVE LOGITS

Through

1.48

 Through

1.45

 Between

1.41

Between

1.38

 Across

1.33

 Around

1.30

Across

1.30

Together

1.27

 Along

1.26

 Against

1.25

Activations Density 0.270%

phrases that express temporal or spatial relationships

No Known Activations