INDEX

Explanations

phrases related to intellectual property and copyright

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-1.44

 intersper

-1.10

 gratify

-1.04

 rouse

-1.02

 fernando

-0.95

 overcrow

-0.94

 endow

-0.92

 quitted

-0.88

 reconno

-0.86

 amass

-0.84

POSITIVE LOGITS

 intellectual

1.32

 Intellectual

1.21

Intellectual

1.16

intellectual

1.15

 intelectual

1.12

lectual

0.91

 intellectually

0.78

 intellectuals

0.73

 intellect

0.71

 withRouter

0.67

Activations Density 0.334%

phrases related to intellectual property and copyright

No Comments

No Known Activations