INDEX

Explanations

references to specific individuals, particularly in legal or formal contexts

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

ness

-0.62

おり

-0.61

ة

-0.60

 שוליים

-0.59

 Pyrene

-0.59

 EconPapers

-0.58

agoza

-0.58

ILayout

-0.58

 eccell

-0.58

 appelez

-0.57

POSITIVE LOGITS

 propOrder

0.74

@"/

0.73

UnsafeEnabled

0.66

✨:

0.65

 numberWith

0.62

ので

0.62

herited

0.62

/*

0.61

mitted

0.59

tinyos

0.57

Activations Density 0.516%

references to specific individuals, particularly in legal or formal contexts

No Comments

No Known Activations