INDEX

Explanations

phrases indicating specific time frames or durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 quebra

-0.35

SequentialGroup

-0.34

StoreMessageInfo

-0.33

ContentLoaded

-0.30

viewBox

-0.30

loc

-0.30

else

-0.29

copyOf

-0.29

Pontos

-0.28

 Absch

-0.27

POSITIVE LOGITS

 while

1.14

 awhile

1.13

while

0.84

WHILE

0.80

 mentre

0.80

 mientras

0.76

 WHILE

0.74

 autorytatywna

0.71

 enquanto

0.71

 considerable

0.71

Activations Density 0.147%

phrases indicating specific time frames or durations

No Comments

No Known Activations

phrases indicating specific time frames or durations

No Comments

No Known Activations