INDEX

Explanations

instances of the word "did"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 itſelf

-1.20

 Theſe

-1.15

 Monfieur

-1.12

 Houſe

-1.06

ſy

-1.03

 Jefus

-1.02

 theſe

-1.02

 houſe

-1.01

 whoſe

-1.01

 pleaſure

-0.99

POSITIVE LOGITS

did

2.20

did

2.03

Did

1.91

Did

1.81

DID

1.80

 didn

1.53

DID

1.53

didn

1.30

 Didn

1.24

had

1.21

Activations Density 0.044%

instances of the word "did"

No Comments

No Known Activations