INDEX

Explanations

phrases or sentences that have been edited or revised with a focus on alteration or improvement

oai_token-act-pair · gpt-3.5-turbo

instances of the word "edited" and its variations

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

umers

-0.63

 Siem

-0.63

eph

-0.62

umer

-0.61

 infrastructure

-0.60

ibaba

-0.60

wart

-0.59

Weak

-0.58

 surface

-0.58

 bandwagon

-0.58

POSITIVE LOGITS

 edited

4.09

edited

2.47

 Edited

2.38

 edits

2.01

 edit

1.95

 editing

1.85

Edited

1.80

 Editing

1.64

 EDIT

1.37

editor

1.28

Activations Density 0.016%

phrases or sentences that have been edited or revised with a focus on alteration or improvement

instances of the word "edited" and its variations

No Comments

No Known Activations