INDEX

Explanations

references to books and reading activities

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

what are you reading

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

UnknownFields

-0.38

jetty

-0.37

rewrite

-0.35

Zunanje

-0.33

ilderness

-0.33

edoria

-0.33

rest

-0.32

Legături

-0.31

numerical

-0.31

 store

-0.31

POSITIVE LOGITS

 iconLine

0.52

UrlResolution

0.51

 referrerpolicy

0.49

saraba

0.48

 AssemblyCulture

0.47

Booked

0.45

AccessorTable

0.45

ajuku

0.45

 pickups

0.44

 trusty

0.44

Activations Density 0.035%

references to books and reading activities

what are you reading

No Comments

No Known Activations