INDEX

Explanations

expressions related to logical proof and demonstration

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

goal to prove

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

tonode

-0.50

erequisite

-0.47

 Scripts

-0.46

Ply

-0.45

хьтан

-0.45

yq

-0.45

enfone

-0.45

utives

-0.44

AER

-0.44

POSITIVE LOGITS

BufferException

0.54

Insee

0.41

hyrchwyd

0.41

addPreferredGap

0.40

تقاوى

0.40

gól

0.39

 kasarigan

0.36

 rapidez

0.36

RTLU

0.36

 whether

0.35

Activations Density 0.078%

expressions related to logical proof and demonstration

goal to prove

No Comments

No Known Activations