INDEX

Explanations

mentions of sports teams, players, and performance statistics

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

djang

-0.62

🕗

-0.53

atguigu

-0.51

funko

-0.50

 transfé

-0.50

 donnera

-0.50

 suspendu

-0.49

 jugé

-0.49

 compréhen

-0.49

zmm

-0.49

POSITIVE LOGITS

 squad

1.21

squad

1.15

 Squad

1.13

Squad

1.05

 squads

1.03

SQU

0.60

 ensemble

0.56

SQU

0.56

 outfit

0.54

 Squadron

0.52

Activations Density 0.104%

mentions of sports teams, players, and performance statistics

No Comments

No Known Activations