INDEX

Explanations

references to individuals in professional contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

LabelTagHelper

-0.56

UrlResolution

-0.54

InjectAttribute

-0.51

UrlEncoded

-0.50

EnumMember

-0.49

 CreateTagHelper

-0.48

hythm

-0.43

 ModelExpression

-0.41

 powering

-0.39

awakeFromNib

-0.39

POSITIVE LOGITS

 served

1.02

 assumed

0.81

 oversaw

0.77

 chaired

0.74

 presided

0.73

 headed

0.72

served

0.70

 overseen

0.70

 resigned

0.69

 supervised

0.68

Activations Density 0.429%

references to individuals in professional contexts

No Known Activations