INDEX

Explanations

reference to important legal concepts or guidelines

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

Followed by numbers or letters

np_acts-logits-general · gemini-2.0-flash

affirmation or factual statements

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

 reportedly

-0.98

 ovviamente

-0.95

 seems

-0.90

Obvious

-0.89

 oczywiście

-0.88

 natuurlijk

-0.88

 seemed

-0.86

 obviously

-0.84

 said

-0.83

 seem

-0.81

POSITIVE LOGITS

 indeed

1.02

 efectivamente

0.90

indeed

0.88

 للمعارف

0.85

 وأن

0.82

 inderdaad

0.79

 actually

0.77

actually

0.73

 ACTUALLY

0.72

 fakty

0.71

Activations Density 1.451%

reference to important legal concepts or guidelines

Followed by numbers or letters

affirmation or factual statements

No Comments

No Known Activations