INDEX

Explanations

negative assertions or denials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 يتيمه

-1.00

DockStyle

-0.88

 Theſe

-0.86

 uſe

-0.86

 myſelf

-0.85

 recto

-0.84

 homoto

-0.82

 Tacitus

-0.80

 purpoſe

-0.80

 ModelExpression

-0.79

POSITIVE LOGITS

is

1.25

was

0.93

è

0.92

isn

0.91

 being

0.91

0.89

WAS

0.88

are

0.84

 becoming

0.84

not

0.83

Activations Density 0.113%

negative assertions or denials

No Comments

No Known Activations