INDEX

Explanations

programming-related constructs and syntax

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

subscription IDs

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

EDEFAULT

-0.54

URLException

-0.50

 enfans

-0.47

stdc

-0.46

 myſelf

-0.45

GEBURTS

-0.44

UrlEncoded

-0.43

 himſelf

-0.42

TMP

-0.42

AccessorTable

-0.41

POSITIVE LOGITS

bibfield

0.40

megen

0.40

klaim

0.37

ganu

0.37

 ब्रेकडाउन

0.36

insic

0.35

 purport

0.35

 resourceCulture

0.34

ppur

0.34

cipar

0.34

Activations Density 0.030%

programming-related constructs and syntax

subscription IDs

No Comments

No Known Activations