INDEX

Explanations

locations and references to physical environments or structures

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

migrationBuilder

-0.34

 fotógrafo

-0.30

 lenker

-0.30

 Haji

-0.30

StringLength

-0.30

 pinulongan

-0.30

selectAll

-0.29

 toros

-0.29

Project

-0.29

Weight

-0.29

POSITIVE LOGITS

 ModelExpression

0.92

 المعيارى

0.80

 surla

0.68

 виправивши

0.65

CppMethod

0.56

 <>",

0.56

 CreateTagHelper

0.55

zzleHttp

0.53

 '\\;'

0.53

 SwitchCompat

0.52

Activations Density 0.994%

locations and references to physical environments or structures

No Comments

No Known Activations