INDEX

Explanations

references to disparities or gaps between groups or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 buiten

-0.53

präsident

-0.52

UTERS

-0.50

一回

-0.47

 corso

-0.45

 suelos

-0.43

 geworden

-0.43

ğu

-0.43

彤

-0.42

utnik

-0.42

POSITIVE LOGITS

gap

1.94

Gap

1.70

gap

1.69

Gap

1.68

 gaps

1.54

 Gaps

1.44

 difference

1.42

GAP

1.39

gaps

1.35

 disparity

1.34

Activations Density 0.768%

references to disparities or gaps between groups or conditions

No Comments

No Known Activations