INDEX

Explanations

specific addresses or locations

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ActionCreators

-0.65

tagHelperRunner

-0.60

WriteTagHelper

-0.57

ngOnInit

-0.51

↹

-0.47

AnimationsModule

-0.45

ViewFeatures

-0.45

httphttps

-0.43

Abp

-0.41

createStatement

-0.41

POSITIVE LOGITS

bootstrapcdn

0.49

spesies

0.48

emplar

0.47

 autorytatywna

0.40

orari

0.39

 Provision

0.39

 kautta

0.38

ungalow

0.38

 Brush

0.38

 shovel

0.37

Activations Density 0.001%

specific addresses or locations

No Comments

No Known Activations