INDEX

Explanations

programming constructs related to function parameters and variable definitions

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

end of code block

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

SharedCtor

-1.05

 мәкал

-0.92

 ujednoznacz

-0.91

expandindo

-0.85

WriteTagHelper

-0.85

 zwiſchen

-0.84

$_=

-0.83

<unused79>

-0.82

[@BOS@]

-0.82

<unused8>

-0.82

POSITIVE LOGITS

0.65

 true

0.52

 false

0.51

false

0.49

true

0.47

 True

0.43

True

0.42

 False

0.41

False

0.38

Activations Density 0.185%

programming constructs related to function parameters and variable definitions

end of code block

No Comments

No Known Activations