INDEX

Explanations

specific names or entities

oai_token-act-pair · gpt-3.5-turbo

specific people and entities associated with events and roles

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

Reviewed

-0.72

wagen

-0.62

achine

-0.57

bilt

-0.56

lde

-0.56

Offline

-0.55

 predec

-0.55

 Vaugh

-0.53

=]

-0.53

 squares

-0.53

POSITIVE LOGITS

Pwr

0.67

ucl

0.59

ucc

0.58

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.58

querque

0.58

pora

0.57

haar

0.51

 Pearl

0.50

 practition

0.49

ãĤ³

0.49

Activations Density 1.366%

specific names or entities

specific people and entities associated with events and roles

No Comments

No Known Activations