INDEX

Explanations

phrases highlighting comparisons or contrasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

ToBounds

-0.90

 Gunn

-0.71

 meningitis

-0.71

 uuidv

-0.67

lid

-0.65

ActivityCompat

-0.65

labelledby

-0.64

 Williamson

-0.64

ظيم

-0.64

 Hass

-0.63

POSITIVE LOGITS

uttosto

1.14

 Rather

0.96

Rather

0.92

rather

0.84

 rather

0.84

Vrij

0.73

THON

0.73

 Fairly

0.71

 Notion

0.71

terly

0.71

Activations Density 0.081%

phrases highlighting comparisons or contrasts

No Comments

No Known Activations