INDEX

Explanations

terms related to investigations and inquiries, particularly in political or legal contexts

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

investigate or investigation

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

InputBorder

-0.60

 للاسماء

-0.59

jsonwebtoken

-0.58

styleType

-0.57

########.

-0.56

 betweenstory

-0.53

 oprot

-0.53

DockStyle

-0.51

AddTagHelper

-0.50

protoimpl

-0.50

POSITIVE LOGITS

 investigate

0.75

 investigates

0.70

 investigação

0.69

 investigating

0.66

 investigation

0.66

 investigated

0.62

 indag

0.60

 investigations

0.59

 investigar

0.57

 Investigate

0.56

Activations Density 0.023%

terms related to investigations and inquiries, particularly in political or legal contexts

investigate or investigation

No Comments

No Known Activations