INDEX

Explanations

instances of punctuation and formatting markers

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

Followed by numbers

np_acts-logits-general · gemini-2.0-flash

multi-lingual words or characters

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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



-0.97

SequentialGroup

-0.89

帖最后由

-0.86

 typelib

-0.79

 propOrder

-0.77

Билгалдахарш

-0.74

StructEnd

-0.73

InjectAttribute

-0.72

:✨

-0.69

ništvo

-0.65

POSITIVE LOGITS

artige

0.69

 solches

0.66

こいつ

0.59

 solche

0.56

 itſelf

0.55

 tää

0.54

 solchen

0.53

ừng

0.52

这张

0.52

 kiin

0.52

Activations Density 0.170%

instances of punctuation and formatting markers

Followed by numbers

multi-lingual words or characters

No Comments

No Known Activations