INDEX

Explanations

professionals and individuals associated with specific fields or activities

oai_token-act-pair · gpt-3.5-turbo

terms associated with various professions and activists

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

âĹ¼

-0.68

aber

-0.68

internal

-0.67

pez

-0.67

imester

-0.67

ciating

-0.67

pring

-0.67

 disparate

-0.66

 bullpen

-0.66

crew

-0.63

POSITIVE LOGITS

 extraord

1.05

Joined

0.93

 Gina

0.80

 Jerome

0.76

 Tony

0.74

 educator

0.71

 Sebast

0.71

 Rupert

0.71

 Dante

0.70

 Ezra

0.70

Activations Density 0.282%

professionals and individuals associated with specific fields or activities

terms associated with various professions and activists

No Known Activations