INDEX

Explanations

phrases indicating capability or possibilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/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.10.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

 purpoſe

-0.82

 Majefty

-0.80

 occaf

-0.80

 pleaſure

-0.79

 houſe

-0.73

 ſtate

-0.70

iscus

-0.68

 fuper

-0.68

QRST

-0.67

 Chriftian

-0.67

POSITIVE LOGITS

 cannot

0.79

 Cannot

0.77

cannot

0.70

Cannot

0.68

Can

0.66

 cant

0.65

can

0.64

 CANNOT

0.64

Can

0.61

can

0.61

Activations Density 0.610%

phrases indicating capability or possibilities

No Comments

No Known Activations