INDEX

Explanations

references to supplementary materials and associated figures in scientific texts

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

numbers, variables, and symbols

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

 leſs

-0.55

tvguidetime

-0.54

 vectorielle

-0.54

 deſt

-0.52

 raiſ

-0.52

 greateſt

-0.51

 uſed

-0.50

 pleaſure

-0.49

 ſtate

-0.49

 Вікі

-0.49

POSITIVE LOGITS

 Meksiku

0.55

__':

0.37

aspect

0.33

 createState

0.32

TabStop

0.32

 ді

0.32

BufferException

0.31

stateMutability

0.31

TableHead

0.31

deps

0.30

Activations Density 0.260%

references to supplementary materials and associated figures in scientific texts

numbers, variables, and symbols

No Comments

No Known Activations