INDEX

Explanations

terms related to specificity and uniqueness

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 algum

-0.57

 alguma

-0.55

 EventArgs

-0.54

is

-0.53

-0.51

ningens

-0.50

if

-0.50

стероид

-0.49

as

-0.49

 huom

-0.49

POSITIVE LOGITS

Diweddarwch

0.76

 myſelf

0.74

 ſeveral

0.71

 Theſe

0.71

 faſt

0.69

 itſelf

0.69

 themſelves

0.69

 Jefus

0.65

 Chriſt

0.65

ῖν

0.65

Activations Density 0.891%

terms related to specificity and uniqueness

No Comments

No Known Activations