INDEX

Explanations

numerical identifiers and references within data-related contexts

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

<start_of_turn>user

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

 prés

-0.49

 nakalista

-0.43

UnusedPrivate

-0.42

 Polskiego

-0.42

-------

-0.40

 fuerzas

-0.39

 déb

-0.38

 témoignage

-0.37

 exceptionnel

-0.37

 internetowa

-0.36

POSITIVE LOGITS

saraba

0.59

errHandler

0.53

 MainAxisSize

0.49

BeginInit

0.47

瘩

0.45

ichio

0.45

 useDispatch

0.45

ibis

0.45

protoc

0.45

bcryptjs

0.45

Activations Density 0.005%

numerical identifiers and references within data-related contexts

<start_of_turn>user

No Comments

No Known Activations