INDEX

Explanations

terms associated with content and involvement in various processes or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 المعيارى

-0.87

RenderAtEndOf

-0.70

ValueStyle

-0.70

ftagPool

-0.69

AntiForgeryToken

-0.61

stdc

-0.58

 використо

-0.58

 يتيمه

-0.57

ian

-0.56

Referințe

-0.56

POSITIVE LOGITS

 both

1.03

 mainly

0.89

 mostly

0.87

 lots

0.82

 only

0.81

 plenty

0.78

 primarily

0.77

 neither

0.76

 several

0.76

 either

0.75

Activations Density 0.631%

terms associated with content and involvement in various processes or activities

No Comments

No Known Activations