INDEX

Explanations

elements related to programming logic and structure

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

code variable assignments and increments

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

ніципалі

-0.60

OGND

-0.57



-0.57

 oprot

-0.51

出版年

-0.49



-0.42

 desmotivaciones

-0.40

localVar

-0.39

allus

-0.38

Autoritní

-0.37

POSITIVE LOGITS

WriteTagHelper

0.47

PYX

0.46

HtmlAttribute

0.45

RectangleBorder

0.40

XmlAccessorType

0.40

ंदीखरीदारी

0.39

awtextra

0.38

 Share

0.36

ophytes

0.36

доми

0.36

Activations Density 0.021%

elements related to programming logic and structure

code variable assignments and increments

No Comments

No Known Activations