INDEX

Explanations

expressions related to sharing, commonality, and collaboration

oai_token-act-pair · gpt-3.5-turbo

references to shared connections or communal relationships

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

paio

-0.91

eeks

-0.75

anwhile

-0.70

ensen

-0.69

nesty

-0.69

ange

-0.68

 desper

-0.68

esa

-0.68

irez

-0.68

evaluate

-0.66

POSITIVE LOGITS

cro

0.87

 shared

0.76

Lis

0.68

 Dunham

0.68

 Shared

0.68

 custody

0.65

 shares

0.65

hare

0.65

 share

0.65

Instance

0.64

Activations Density 0.020%

expressions related to sharing, commonality, and collaboration

references to shared connections or communal relationships

No Comments

No Known Activations