INDEX

Explanations

references to previous research, studies, or investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

TypedDataSet

-0.60

moduleName

-0.57

MultipartFile

-0.56

Hentet

-0.55

cticut

-0.54

ArgsConstructor

-0.53

 themſelves

-0.50

MessageTagHelper

-0.50

AutoField

-0.49

IntoConstraints

-0.49

POSITIVE LOGITS

 previous

1.20

Previous

1.15

 Previous

1.15

 previously

1.12

previous

1.10

 PREVIOUS

1.08

PREVIOUS

0.98

 earlier

0.95

previously

0.93

 Previously

0.92

Activations Density 0.142%

references to previous research, studies, or investigations

No Comments

No Known Activations