INDEX

Explanations

technical terms and concepts related to parameters, measurements, and configurations

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

parameter maps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

Meta

-0.33

Abstract

-0.32

this

-0.31

First

-0.30

 Wild

-0.30

Wild

-0.30

prendre

-0.30

Wiel

-0.30

 Danh

-0.30

POSITIVE LOGITS

 ujednoznacz

0.94

 Chwiliwch

0.93

ThroughAttribute

0.91

 EconPapers

0.81

 ProtoMessage

0.81

WriteTagHelper

0.80

fjspx

0.79

findpost

0.77

 ModelExpression

0.76

msgTypes

0.75

Activations Density 0.444%

technical terms and concepts related to parameters, measurements, and configurations

parameter maps

No Comments

No Known Activations