INDEX

Explanations

intensifiers that convey strong emotions or opinions

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

Datuak

-1.02

 linkovi

-0.77

 Martine

-0.73

 Shades

-0.73

NOC

-0.71

Pozdrawiam

-0.71

uset

-0.71

Shades

-0.70

rsiniz

-0.70

MatButtonModule

-0.69

POSITIVE LOGITS

 REALLY

1.05

 really

1.03

 realy

0.94

 Really

0.93

really

0.92

ibatis

0.88

benar

0.84

Really

0.80

 quite

0.75

uttosto

0.64

Activations Density 0.048%

intensifiers that convey strong emotions or opinions

No Comments

No Known Activations