INDEX

Explanations

phrases related to questions or statements indicating surprise or inquiry

oai_token-act-pair · gpt-3.5-turbo

forms of the verb "to be," indicating states or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Dragonbound

-0.61

 Balance

-0.58

otic

-0.58

 balance

-0.56

 cliffs

-0.56

 resonance

-0.55

aucus

-0.55

 fixation

-0.55

ONS

-0.55

stood

-0.55

POSITIVE LOGITS

wolves

1.15

wolf

1.02

abi

0.99

hes

0.96

ps

0.86

nesday

0.86

sbm

0.83

hers

0.83

iago

0.77

abase

0.77

Activations Density 0.073%

phrases related to questions or statements indicating surprise or inquiry

forms of the verb "to be," indicating states or conditions

No Comments

No Known Activations