INDEX

Explanations

phrases and terms related to news broadcasts and reporting structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

TagMode

-0.79

twimg

-0.78

setVerticalGroup

-0.74

bootstrapcdn

-0.73

 AssemblyCompany

-0.70

 CreateTagHelper

-0.69

WaitGroup

-0.68

 Савезне

-0.67

#+#

-0.67

SourceChecksum

-0.67

POSITIVE LOGITS

 occasional

0.74

 occasionally

0.69

 some

0.65

 Occasionally

0.64

Occasionally

0.60

 option

0.60

ExceptionHandler

0.58

 sometimes

0.58

 Occasional

0.57

有一些

0.57

Activations Density 0.568%

phrases and terms related to news broadcasts and reporting structures

No Comments

No Known Activations