INDEX

Explanations

phrases indicating necessity or obligation

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate items or recommendations that are essential or highly suggested

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Ips

-0.67

 Reloaded

-0.66

 twisted

-0.63

 Tempest

-0.63

 Bohem

-0.62

 proprietary

-0.62

 Totem

-0.62

 tort

-0.61

ãĤ§

-0.61

quickShipAvailable

-0.60

POSITIVE LOGITS

reads

1.22

have

1.11

read

1.06

dos

1.03

do

0.98

list

0.97

know

0.96

buy

0.96

answer

0.94

kept

0.92

Activations Density 0.023%

phrases indicating necessity or obligation

phrases that indicate items or recommendations that are essential or highly suggested

No Comments

No Known Activations