INDEX

Explanations

numbers related to votes in a listing or poll

oai_token-act-pair · gpt-3.5-turbo

numerical ratings or scores associated with various items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

©¶æ¥µ

-0.95

-0.88

¿½

-0.88

ailability

-0.86

ŃĶ

-0.84

ccording

-0.84

Pages

-0.83

IPS

-0.83

Sty

-0.82

Artist

-0.82

POSITIVE LOGITS

win

0.86

 skip

0.80

 bill

0.78

min

0.78

 leave

0.77

 loop

0.76

 bonus

0.76

 error

0.76

 plant

0.75

 minute

0.74

Activations Density 0.091%

numbers related to votes in a listing or poll

numerical ratings or scores associated with various items

No Comments

No Known Activations