INDEX

Explanations

negative descriptions directed at a specific person

oai_token-act-pair · gpt-3.5-turbo

negative attributes or criticisms associated with entities or situations

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

 bounce

-0.87

 square

-0.77

 drawer

-0.74

 hiber

-0.74

 drink

-0.73

 fright

-0.71

 breathing

-0.70

sep

-0.69

 context

-0.69

 breat

-0.69

POSITIVE LOGITS

FM

1.38

LP

1.22

derived

1.21

II

1.20

III

1.19

DL

1.16

SN

1.15

RN

1.14

related

1.13

induced

1.12

Activations Density 0.067%

negative descriptions directed at a specific person

negative attributes or criticisms associated with entities or situations

No Comments

No Known Activations