INDEX

Explanations

information targeting the reader and providing advice or instructions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-3.48

ⓧ

-0.95

 onBind

-0.82

EndProject

-0.80

HasIndex

-0.80

Aholisi

-0.75

 AssemblyCompany

-0.73

ContentAsync

-0.71

omock

-0.69

rrggbb

-0.68

POSITIVE LOGITS

 maneu

1.94

 reluct

1.94

 impra

1.82

 disagre

1.79

 affor

1.75

 accla

1.75

 inev

1.73

 volunte

1.71

 increa

1.70

 apprehen

1.67

Activations Density 0.259%

information targeting the reader and providing advice or instructions

No Comments

No Known Activations