INDEX

Explanations

terminology related to mid-sized concepts or themes

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

']?>

-0.78

 uVar

-0.78

таратура

-0.78

')]

-0.76

 Tavares

-0.73

ので

-0.72

%)$

-0.71

InputBorder

-0.70

)])

-0.69

']}

-0.68

POSITIVE LOGITS

MID

1.81

Mid

1.74

Mid

1.73

mid

1.72

mid

1.60

MID

1.59

 mids

1.25

 Midway

1.23

 midwives

1.22

 Middel

1.21

Activations Density 0.046%

terminology related to mid-sized concepts or themes

No Comments

No Known Activations