INDEX

Explanations

code-related actions or commands

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

Code snippets or code-related terms

np_acts-logits-general · gemini-2.0-flash

code syntax elements

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

NamedQueries

-0.76

mybatisplus

-0.71

+#+

-0.69

ModelAdmin

-0.61

WebServlet

-0.60

RectangleBorder

-0.57

 ब्रेकडाउन

-0.57

 myſelf

-0.57

RegressionTest

-0.56

 anſ

-0.55

POSITIVE LOGITS

 Numerade

0.45

 nakalista

0.40

 ivelany

0.40

PreAuthorize

0.36

 leading

0.35

 announces

0.34

 объ

0.34

 series

0.34

 announce

0.33

 Announced

0.33

Activations Density 0.550%

code-related actions or commands

Code snippets or code-related terms

code syntax elements

No Comments

No Known Activations