INDEX

Explanations

mentions of medical modifications or procedures

oai_token-act-pair · gpt-3.5-turbo

variations of the word "mod" indicating modified entities or concepts

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

ģĸ

-0.80

ï¸

-0.76

 Lauder

-0.73

izabeth

-0.72

terday

-0.69

UGE

-0.66

âĸ¬

-0.64

FORE

-0.64

OHN

-0.64

 Vaugh

-0.64

POSITIVE LOGITS

ifiers

1.23

ifiable

1.21

ifications

1.21

ding

1.18

icum

1.18

ifier

1.13

elled

1.11

ded

1.09

ifying

1.08

ifies

1.08

Activations Density 0.011%

mentions of medical modifications or procedures

variations of the word "mod" indicating modified entities or concepts

No Comments

No Known Activations