INDEX

Explanations

phrases indicating the presence or existence of content or information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 Orville

-0.82

-0.65

 nhau

-0.64

lüğ

-0.64

DPP

-0.62

 userService

-0.61

 @"";

-0.60

bigskip

-0.60

 років

-0.58

 Leona

-0.57

POSITIVE LOGITS

 Contain

1.41

 CONTAIN

1.41

 contains

1.35

 contain

1.32

 Contains

1.28

 contained

1.27

 enthalten

1.24

 Containing

1.14

contains

1.13

Contain

1.13

Activations Density 0.089%

phrases indicating the presence or existence of content or information

No Comments

No Known Activations