INDEX

Explanations

instances of the word "inform" and its variations, indicating a focus on communication or notification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arithmic

-0.58

oherty

-0.56

amins

-0.55

igraphic

-0.55

umoj

-0.54

endgroup

-0.54

glLoadIdentity

-0.54

IDL

-0.54

erializer

-0.53

popularity

-0.53

POSITIVE LOGITS

 tell

1.28

Tell

1.28

 Tell

1.23

 Telling

1.22

 tells

1.22

 Tells

1.19

 telling

1.18

 TELL

1.17

Telling

1.14

TELL

1.08

Activations Density 0.505%

instances of the word "inform" and its variations, indicating a focus on communication or notification

No Comments

No Known Activations