INDEX

Explanations

patterns related to programming code, specifically involving formatting and variables

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

 sculptured

-0.71

 vexed

-0.71

 unspeak

-0.70

 McLaugh

-0.69

 savag

-0.64

 heapq

-0.63

 pensive

-0.63

 woodcut

-0.63

 languid

-0.62

 cauldron

-0.62

POSITIVE LOGITS

 Ottobre

1.27

 Settembre

1.16

 incess

1.12

 Giugno

1.11

 pecuni

1.09

 sappi

1.08

 Luglio

1.05

 solidar

1.02

 apparti

1.02

 cappello

1.02

Activations Density 0.042%

patterns related to programming code, specifically involving formatting and variables

No Comments

No Known Activations