INDEX

Explanations

computer programming terms and code constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 الحره

-0.91

ImageContext

-0.90

 Normdatei

-0.83

хьтан

-0.79

 الرياضيه

-0.76

 pinulongan

-0.74

AutoScaleMode

-0.74

RuleContext

-0.72

出版年

-0.70

 AssemblyCulture

-0.67

POSITIVE LOGITS

flashdata

0.55

bladet

0.47

roxene

0.47

haup

0.47

그

0.46

pats

0.46

pyplot

0.44

newArrayList

0.44

arikat

0.44

bean

0.43

Activations Density 0.243%

computer programming terms and code constructs

No Comments

No Known Activations