INDEX

Explanations

phrases related to addressing issues or concerns, specifically in a context that involves making statements or taking action to resolve them

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

 compen

-0.75

 suscep

-0.73

 unden

-0.72

 disreg

-0.69

 excru

-0.69

 uninten

-0.68

 opport

-0.68

 reluct

-0.68

 inev

-0.66

 benevol

-0.66

POSITIVE LOGITS

 address

1.16

 addressed

1.13

 addresses

1.12

 addressing

1.09

 Address

1.06

address

1.03

addressed

1.00

 ADDRESS

1.00

ADDRESS

0.97

Address

0.94

Activations Density 0.064%

phrases related to addressing issues or concerns, specifically in a context that involves making statements or taking action to resolve them

No Comments

No Known Activations