INDEX

Explanations

specific terms related to assessments and classifications in various contexts

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

allows, enables, makes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

__':

-0.62

__':

-0.50

__":

-0.49

__":

-0.47

PreferredItem

-0.46

jfree

-0.46

 resourceCulture

-0.44

Hentet

-0.43

[]:

-0.42

TextAppearance

-0.42

POSITIVE LOGITS

 ensures

0.47

 allows

0.45

 helps

0.43

 makes

0.40

씩

0.38

 prevents

0.38

enable

0.37

web

0.37

 results

0.36

 cassa

0.36

Activations Density 0.217%

specific terms related to assessments and classifications in various contexts

allows, enables, makes

No Comments

No Known Activations