INDEX

Explanations

references to technical processes and components

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

said the

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

 Autorisations

-0.46

rungsseite

-0.44

 nonUne

-0.43

 revisor

-0.43

 charities

-0.41

httphttps

-0.41

outfile

-0.41

帖最后由

-0.41



-0.40

myfile

-0.40

POSITIVE LOGITS

 aforementioned

0.62

 aforesaid

0.61

Said

0.57

 said

0.54

 citada

0.54

said

0.53

 appunto

0.50

canestro

0.50

 tersebut

0.48

 Said

0.48

Activations Density 0.288%

references to technical processes and components

said the

No Comments

No Known Activations