INDEX

Explanations

locations or directions indicated by prepositions like 'under' or 'underneath'

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Argumento

-0.59

hdashline

-0.50

Escolha

-0.48

keyboardType

-0.48

GOTREF

-0.48

%)$

-0.47

allclose

-0.47

vokat

-0.47

BuildContext

-0.47

DropDownList

-0.45

POSITIVE LOGITS

 UNDER

0.80

 jette

0.79

aen

0.77

 Gorb

0.76

 accompagne

0.76

hcm

0.75

fta

0.75

 Beneath

0.74

 marié

0.74

 beneath

0.73

Activations Density 0.074%

locations or directions indicated by prepositions like 'under' or 'underneath'

No Comments

No Known Activations