INDEX

Explanations

variable names or abbreviations related to biological and computational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

:✨

-0.83

PerformLayout

-0.73

GEBURTSDATUM

-0.73

Diwedd

-0.73

 nahilalakip

-0.71

 "..\..\..\

-0.64

Portail

-0.63

awaiter

-0.63

RenderAtEndOf

-0.63

---*/

-0.62

POSITIVE LOGITS

ggg

0.93

ging

0.92

gggg

0.91

Kag

0.81

ged

0.78

ges

0.76

gg

0.76

gy

0.75

SAG

0.74

Cag

0.73

Activations Density 1.102%

variable names or abbreviations related to biological and computational contexts

No Comments

No Known Activations