INDEX

Explanations

phrases expressing uncertainty 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 @ 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

istic

-0.53

’

-0.50

memberof

-0.47

-0.43

menuStrip

-0.43

 numele

-0.43

ism

-0.43

-0.42

ATE

-0.42

️

-0.42

POSITIVE LOGITS

verwijspagina

0.95

PerformLayout

0.90

 WebDriverWait

0.82

AndroidJUnit

0.82

lessly

0.81

ingly

0.80

 CreateTagHelper

0.78

 destined

0.78

 >=",

0.77

 defaultstate

0.76

Activations Density 0.077%

phrases expressing uncertainty or opinions

No Comments

No Known Activations