INDEX

Explanations

references to job titles or roles

oai_token-act-pair · gpt-3.5-turbo

key items and important dates or events

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

etheless

-0.72

quartered

-0.69

isode

-0.69

Moore

-0.56

 Vaugh

-0.56

Topics

-0.56

opped

-0.55

izabeth

-0.55

FTWARE

-0.54

ascular

-0.54

POSITIVE LOGITS

 reader

0.76

lang

0.72

 Quan

0.68

ulhu

0.63

 Scarlet

0.61

 Pastebin

0.61

tch

0.60

 Beef

0.58

gz

0.58

ows

0.58

Activations Density 1.064%

references to job titles or roles

key items and important dates or events

No Comments

No Known Activations