INDEX

Explanations

negations and phrases expressing disagreement or urging caution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

UniformLocation

-0.58

 ProtoMessage

-0.57

Tracce

-0.57

شوف

-0.48

 aikaa

-0.47

 pylint

-0.46

 aapt

-0.46

 đích

-0.46

 calendriers

-0.45

istolet

-0.45

POSITIVE LOGITS

 worry

1.07

 forget

1.05

 Worry

0.81

 underestimate

0.78

 Forget

0.76

 fret

0.75

worry

0.75

 waste

0.73

Forget

0.71

 оригіналу

0.71

Activations Density 0.103%

negations and phrases expressing disagreement or urging caution

No Comments

No Known Activations