INDEX

Explanations

words or phrases related to programming/coding, such as variable names and code snippets

oai_token-act-pair · gpt-3.5-turbo

numerical values and quantities 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 treaty

-0.76

 covenant

-0.69

 grop

-0.67

pro

-0.66

 unconditional

-0.65

 sculpt

-0.64

 treaties

-0.64

Mub

-0.64

 freedom

-0.63

 constitutionally

-0.63

POSITIVE LOGITS

RAW

1.27

Advertisements

1.24

NAS

1.17

Reviewer

1.16

Related

1.05

Page

1.03

Updated

1.03

Posted

1.03

Topics

1.02

Advertisement

0.99

Activations Density 0.497%

words or phrases related to programming/coding, such as variable names and code snippets

numerical values and quantities in various contexts

No Comments

No Known Activations