INDEX

Explanations

distinct markers or indicators of a structured data environment

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

Punctuation and symbols

np_acts-logits-general · gemini-2.0-flash

research papers or academic articles

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

).\\

-0.86

}}$\\

-0.79

;\\

-0.78

?\\

-0.73

########.

-0.70

WriteAttribute

-0.68

Glej

-0.68

:\\

-0.66

.\\

-0.65

POSITIVE LOGITS

</em>

1.11

</strong>

1.02

</u>

0.96

”

0.83

</h5>

0.81

。”

0.78

.”

0.76

?”

0.73

</h4>

0.70

</i>

0.69

Activations Density 0.012%

distinct markers or indicators of a structured data environment

Punctuation and symbols

research papers or academic articles

No Comments

No Known Activations