INDEX

Explanations

references to groups of people, particularly in the context of social or health-related discussions

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

-0.84

-0.66

、

-0.64

-,

-0.58

 serta

-0.58

-0.57

-0.54

以及

-0.53

／

-0.53

POSITIVE LOGITS

 Мексичка

1.30

########.

1.28

GEBURTSDATUM

1.17

findpost

1.17

 تانيه

1.16

HtmlAttribute

1.16

RenderAtEndOf

1.16

 propOrder

1.13

 дописавши

1.11

Билгалдахарш

1.10

Activations Density 0.647%

references to groups of people, particularly in the context of social or health-related discussions

No Comments

No Known Activations