INDEX

Explanations

special markers or tokens indicating the beginning or structure of sections within a document

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

Rosko

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

iettes

-0.50

дента

-0.46

tonode

-0.46

 nyelven

-0.46

 cell

-0.46

</thead>

-0.44

one

-0.43

expérience

-0.41

gap

-0.41

databind

-0.41

POSITIVE LOGITS

########.

1.03

 فريبيس

0.98

 ModelExpression

0.93

SourceChecksum

0.89

 HasFactory

0.84

发表于

0.82

 صوتيه

0.82

IVEREF

0.78

RenderAtEndOf

0.77

 estekak

0.77

Activations Density 0.582%

special markers or tokens indicating the beginning or structure of sections within a document

Rosko

No Comments

No Known Activations