INDEX

Explanations

references to triviality or insignificance in various contexts

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

ctivité

-0.50

żeń

-0.49

 longest

-0.49

nesia

-0.49

dedicated

-0.48

SPATH

-0.48

mad

-0.47

 armes

-0.47

 mema

-0.47

ươi

-0.47

POSITIVE LOGITS

 trivial

1.12

 trivi

1.01

 minor

0.98

trivial

0.89

minor

0.88

 insignificant

0.88

 unimportant

0.88

 мело

0.84

 <<<<<<<<<<<<<<

0.83

 betweenstory

0.81

Activations Density 0.290%

references to triviality or insignificance in various contexts

No Comments

No Known Activations

references to triviality or insignificance in various contexts

No Comments

No Known Activations