INDEX

Explanations

connections between information or data points within a system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

strokeWeight

-0.50

enumi

-0.47

 country

-0.47

kega

-0.46

riet

-0.46

 Nether

-0.42

GeneratedValue

-0.42

makeText

-0.41

woff

-0.41

country

-0.40

POSITIVE LOGITS

StoryboardSegue

0.82

ніципалі

0.77

ثيق

0.76

 unknownFields

0.70

 itself

0.70

itself

0.65

 trajets

0.62

 تضيفلها

0.61

OfWork

0.60

 próprias

0.59

Activations Density 1.540%

connections between information or data points within a system

No Comments

No Known Activations