INDEX

Explanations

words related to errors or mistakes

oai_token-act-pair · gpt-3.5-turbo

mentions of various types of errors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Stre

-0.77

cape

-0.73

chin

-0.68

ques

-0.68

æ©

-0.67

ofi

-0.66

 domestically

-0.66

kefeller

-0.64

bush

-0.64

adelphia

-0.64

POSITIVE LOGITS

 error

4.03

 errors

2.86

 Error

2.73

error

2.51

Error

2.25

 mistake

2.20

 Errors

2.17

 ERROR

1.88

errors

1.84

 mistakes

1.78

Activations Density 0.027%

words related to errors or mistakes

mentions of various types of errors

No Comments

No Known Activations