INDEX

Explanations

words related to reviewing or critiquing various forms of media or art

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

Geographie

-0.47

 iyon

-0.45

loride

-0.44

 habet

-0.44

 mús

-0.43

 getStart

-0.41

viewWillAppear

-0.41

 vost

-0.41

Composable

-0.40

FloatField

-0.40

POSITIVE LOGITS

 EVER

0.91

 ever

0.86

 Ever

0.76

EVER

0.75

Ever

0.73

ⓧ

0.68

ever

0.68

 dichi

0.65

 gardien

0.63

 jamás

0.62

Activations Density 0.065%

words related to reviewing or critiquing various forms of media or art

No Comments

No Known Activations