INDEX

Explanations

words and phrases related to products or services being considered as the best or recommended in a specific context

oai_token-act-pair · gpt-3.5-turbo

references to optimal choices or the best options in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rocal

-0.74

 contag

-0.73

utory

-0.67

ilial

-0.66

 illeg

-0.64

orate

-0.64

ogenesis

-0.64

ormonal

-0.63

olid

-0.63

 aggravated

-0.62

POSITIVE LOGITS

 EVER

0.82

 congr

0.77

 compliment

0.77

Score

0.70

 Choice

0.68

 unbeat

0.67

 yours

0.66

 unbiased

0.66

expensive

0.66

 shortest

0.66

Activations Density 0.458%

words and phrases related to products or services being considered as the best or recommended in a specific context

references to optimal choices or the best options in various contexts

No Comments

No Known Activations