INDEX

Explanations

phrases related to gaps or discrepancies

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

 fortn

-0.84

 compen

-0.83

 encomp

-0.82

oun

-0.79

 Ruman

-0.79

Hn

-0.76

 oner

-0.75

 alre

-0.75

 attemp

-0.74

 Tole

-0.73

POSITIVE LOGITS

gap

1.38

gap

1.30

 gaps

1.19

Gap

1.16

Gap

1.13

GAP

0.93

gaps

0.92

 Gaps

0.83

GAP

0.78

 between

0.72

Activations Density 0.105%

phrases related to gaps or discrepancies

No Comments

No Known Activations