INDEX

Explanations

commands and actions related to typing or entering information

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

computer-related instructions such as commands, typing, and entering information

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reilly

-0.68

arers

-0.68

 reorgan

-0.67

 exhibiting

-0.65

 exhibition

-0.61

icable

-0.59

¥µ

-0.59

gas

-0.59

 feud

-0.59

otics

-0.58

POSITIVE LOGITS

 captcha

1.00

ername

0.96

 passwords

0.94

codes

0.90

 NAME

0.87

"%

0.87

 username

0.86

 commands

0.84

 initials

0.82

 words

0.80

Activations Density 0.109%

commands and actions related to typing or entering information

computer-related instructions such as commands, typing, and entering information

No Comments

No Known Activations