INDEX

Explanations

punctuation marks and emoticons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/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.35.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

ряду

-0.36

 FormData

-0.34

bibitem

-0.32

 seguimiento

-0.32

mitian

-0.32

ligente

-0.32

 gild

-0.31

 Famous

-0.31

 mỡ

-0.30

chymal

-0.30

POSITIVE LOGITS

!:

0.73

::::::::::::::::

0.71

::::

0.69

::::::::

0.68

:::

0.68

|:

0.64

/:

0.60

~:

0.58

...:

0.57

!!:

0.55

Activations Density 0.282%

punctuation marks and emoticons

No Comments

No Known Activations