INDEX

Explanations

phrases that indicate evaluation or consideration of a subject

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

 haine

-0.70

 brevity

-0.70

igert

-0.69

 ModelExpression

-0.69

 Enhanced

-0.68

 explication

-0.66

Enhanced

-0.65

setLayoutManager

-0.65

 Tada

-0.63

enhanced

-0.62

POSITIVE LOGITS

 considered

1.45

considered

1.29

 Considered

1.21

 considerado

1.13

 regarded

1.03

 considerada

1.00

 consideradas

1.00

idered

0.99

 deemed

0.99

 considerados

0.99

Activations Density 0.215%

phrases that indicate evaluation or consideration of a subject

No Comments

No Known Activations