INDEX

Explanations

terms related to specific practices or actions being discussed, particularly in a professional context

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

 congé

-0.47

 nomme

-0.44

Cdt

-0.44

veu

-0.44

Infl

-0.44

 évidence

-0.43

FDR

-0.43

 Waff

-0.43

 considération

-0.43

 conformité

-0.42

POSITIVE LOGITS

 practice

1.24

 practices

1.21

practice

1.20

 Practice

1.16

Practice

1.14

 Practices

1.13

practices

1.11

 practise

1.06

 PRACTICE

1.04

 practiced

1.00

Activations Density 0.070%

terms related to specific practices or actions being discussed, particularly in a professional context

No Comments

No Known Activations