INDEX

Explanations

Ways of indicating or reacting to specific incidents, actions, or events

oai_token-act-pair · gpt-3.5-turbo

events or situations involving video evidence or visual documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ntil

-0.82

DragonMagazine

-0.69

soDeliveryDate

-0.67

Topic

-0.65

 Cosponsors

-0.64

UGC

-0.62

ãĥ¼ãĤ¯

-0.62

ETF

-0.62

LESS

-0.61

NetMessage

-0.60

POSITIVE LOGITS

 arrives

0.79

 expires

0.70

 arrived

0.69

 erupted

0.69

 arrive

0.68

 abruptly

0.68

 uproar

0.67

 outcry

0.67

 enters

0.65

 unexpectedly

0.65

Activations Density 0.991%

Ways of indicating or reacting to specific incidents, actions, or events

events or situations involving video evidence or visual documentation

No Comments

No Known Activations