INDEX

Explanations

phrases related to events happening in the past and present

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a subsequent update or change over time

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

assed

-0.69

anium

-0.61

 Tick

-0.60

 Vand

-0.59

toe

-0.59

 Syndicate

-0.59

venture

-0.58

 centerpiece

-0.58

hab

-0.56

mson

-0.56

POSITIVE LOGITS

forth

0.81

DragonMagazine

0.80

--------------------------------

0.76

SPONSORED

0.75

Reviewer

0.73

abouts

0.70

————————————————

0.69

èª

0.68

================

0.68

å£«

0.68

Activations Density 0.036%

phrases related to events happening in the past and present

phrases that indicate a subsequent update or change over time

No Comments

No Known Activations