INDEX

Explanations

programming-related terms and conditions

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

code block delimiters like ````

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

 Ireland

-0.44

$}}

-0.42

 Bour

-0.41

Bok

-0.40

Ig

-0.40

IBOutlet

-0.40

PEP

-0.40

Jä

-0.39

 BOUR

-0.39

Bour

-0.39

POSITIVE LOGITS

mybatisplus

0.64

PreExecute

0.60

 mxArray

0.54

 nakalista

0.52

VYMaps

0.52

Tikang

0.49

SerializedSize

0.49

Atsauces

0.48

TagNumber

0.48

niająca

0.48

Activations Density 0.005%

programming-related terms and conditions

code block delimiters like ````

No Comments

No Known Activations