INDEX

Explanations

terms related to severity and harshness

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-2.32

DataPropertyName

-0.69

 reconno

-0.68

 supposes

-0.63

 vainly

-0.61

glMatrixMode

-0.61

 enshr

-0.61

 recollect

-0.61

 quitted

-0.60

 bring

-0.60

POSITIVE LOGITS

 palio

1.46

 ristor

1.34

 paradiso

1.32

 maroc

1.31

 bandung

1.31

 tramont

1.27

 riva

1.24

capulco

1.22

 toscana

1.20

 monaster

1.19

Activations Density 0.129%

terms related to severity and harshness

No Comments

No Known Activations