INDEX

Explanations

terms related to mutations and their variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 Dede

-0.75

flink

-0.70

 createSprite

-0.64

 infor

-0.64

antel

-0.62

<blockquote>

-0.62

 paff

-0.62

 Steg

-0.62

Kog

-0.61

 rind

-0.61

POSITIVE LOGITS

Mu

1.47

Mut

1.29

MUT

1.24

Mu

1.23

 mutate

1.22

Mut

1.20

mu

1.19

mu

1.19

MU

1.19

mut

1.14

Activations Density 0.356%

terms related to mutations and their variants

No Comments

No Known Activations