INDEX

Explanations

references indicating alternative options or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

charCode

-0.58

ulemon

-0.56

<?

-0.50

uminescence

-0.49

initState

-0.45

 Benz

-0.45

transQ

-0.43

IMP

-0.43

 transfieras

-0.42

VID

-0.42

POSITIVE LOGITS

 OTHER

1.13

 other

1.11

 Other

1.09

other

1.08

Other

1.05

 Autres

0.96

 Otras

0.94

OTHER

0.94

autres

0.89

Autres

0.87

Activations Density 0.171%

references indicating alternative options or comparisons

No Comments

No Known Activations