INDEX

Explanations

patterns of numerical sequences or references to legal citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

nloa

-0.51

 onBind

-0.47

wpi

-0.46

 orta

-0.45

ITC

-0.44

 lowland

-0.44

 southeastern

-0.44

 obes

-0.43

hac

-0.43

imine

-0.43

POSITIVE LOGITS

 nahilalakip

0.75

ьаж

0.66

IBOutlet

0.62

ztály

0.61

OMON

0.61

ESSOR

0.60

eynman

0.60

 Familienname

0.58

TypedDataSet

0.58

MemoryWarning

0.55

Activations Density 0.601%

patterns of numerical sequences or references to legal citations

No Comments

No Known Activations