INDEX

Explanations

words related to extreme scenarios or emotions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 philanth

-0.80

 exorbit

-0.75

 horrend

-0.71

 shenan

-0.70

 Birgit

-0.67

 compréhen

-0.67

 dilap

-0.66

 Karsten

-0.66

 cuck

-0.66

 chauffe

-0.65

POSITIVE LOGITS

 sentito

0.68

 parlato

0.64

isContained

0.64

 chiesto

0.63

 DISEASES

0.62

 provato

0.62

FERENCE

0.61

SCRIBE

0.59

 potuto

0.59

VISIONS

0.57

Activations Density 0.410%

words related to extreme scenarios or emotions

No Comments

No Known Activations