INDEX

Explanations

words related to design and composition processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

<eos>

-0.64

it

-0.56

not

-0.54

ter

-0.52

now

-0.51

-0.49

per

-0.47

can

-0.46

its

-0.46

><!--

-0.46

POSITIVE LOGITS

 myſelf

1.34

 itſelf

1.22

 becauſe

1.19

 Monfieur

1.18

 ―――――

1.16

 Theſe

1.14

 Anſ

1.12

 ſmall

1.12

Efq

1.11

 againſt

1.11

Activations Density 0.403%

words related to design and composition processes

No Comments

No Known Activations