INDEX

Explanations

expressions indicating safety or security

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.10.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

chanind/openwebtext-gemma

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

ctigges/openwebtext-gemma-1024-cl

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-2.82

-0.74

/**

-0.72

<?

-0.71

 exert

-0.63

ⓧ

-0.62

 rehabilitate

-0.62

 reunite

-0.61

 accumulate

-0.60

 realize

-0.58

POSITIVE LOGITS

cæ

1.35

 magis

1.30

 socie

1.25

 Bibl

1.22

 nomine

1.19

 sentra

1.17

 tramont

1.17

 casio

1.16

 megane

1.16

 milano

1.16

Activations Density 0.038%

expressions indicating safety or security

No Comments

No Known Activations