INDEX

Explanations

names of people or characters

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names associated with legislation and characters

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

anwhile

-0.65

jri

-0.64

eret

-0.58

ascript

-0.57

ruary

-0.52

gotten

-0.51

 notor

-0.51

iqueness

-0.50

 predec

-0.50

 sacrific

-0.49

POSITIVE LOGITS

á

0.54

 ..........

0.53

 Tenth

0.47

è£ħ

0.47

 Archdemon

0.46

 Miko

0.45

ãĤ¼ãĤ¦ãĤ¹

0.45

hua

0.45

Dest

0.44

 Mons

0.44

Activations Density 1.044%

names of people or characters

proper nouns and names associated with legislation and characters

No Comments

No Known Activations