INDEX

Explanations

numeric sequences and formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 nahilalakip

-1.14

Efq

-0.96

TintMode

-0.81

 ModelExpression

-0.77

 Monfieur

-0.77

tvguidetime

-0.77

 Jefus

-0.76

 Majefty

-0.74

Datuak

-0.73

 tartalomajánló

-0.72

POSITIVE LOGITS

 Drapeau

0.52

neb

0.49

 come

0.47

 volverá

0.46

nani

0.44

 parc

0.43

able

0.42

 tutt

0.42

散

0.42

lossians

0.42

Activations Density 0.303%

numeric sequences and formatting

No Comments

No Known Activations