INDEX

Explanations

legal terminology related to court cases and defendants

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

parameter partition

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

inalg

-0.31

colgroup

-0.29

■

-0.27

收

-0.27

mær

-0.27

Cip

-0.27

：(

-0.25

Due

-0.25

Web

-0.24

 argc

-0.24

POSITIVE LOGITS

PathVariable

5.81

PathParam

1.88

RequestParam

1.45

 AssemblyCompany

0.99

 AssemblyTitle

0.96

ModelAttribute

0.90

RequestBody

0.87

 autorytatywna

0.84

 EconPapers

0.81

featureID

0.80

Activations Density 0.000%

legal terminology related to court cases and defendants

parameter partition

No Comments

No Known Activations