INDEX

Explanations

elements in a structured document, specifically focusing on the presence of metadata and formatting details

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

following "the"

np_acts-logits-general · gemini-2.0-flash

the followed by article/item

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

]")]

-0.70

chmal

-0.60

__':

-0.60

 maxn

-0.59

 محفوظة

-0.59

\{\\

-0.58

 fVar

-0.58

 AppColors

-0.57

kegaard

-0.56

 MessageType

-0.56

POSITIVE LOGITS

 following

0.89

 corrispond

0.78

 appropriate

0.73

following

0.73

 default

0.73

 last

0.73

 above

0.72

 offending

0.68

 below

0.67

0.66

Activations Density 0.631%

elements in a structured document, specifically focusing on the presence of metadata and formatting details

following "the"

the followed by article/item

No Comments

No Known Activations