INDEX

Explanations

words likely to be near the end of sentences

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

preceding "for"

np_acts-logits-general · gemini-2.0-flash

end of sentence nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

-1.19

 Мексичка

-1.15

__':

-1.15

++

-1.13

?')

-1.10

 كومونز

-1.09

[]

-1.08

)";

-1.08

$")

-1.07

[])

-1.04

POSITIVE LOGITS

3.03

).

1.51

。

1.49

."

1.31

".

1.31

.”

1.30

1.26

}$.

1.24

].

1.24

”.

1.24

Activations Density 13.172%

words likely to be near the end of sentences

preceding "for"

end of sentence nouns

No Comments

No Known Activations