INDEX

Explanations

titles, announcements, and events

oai_token-act-pair · gpt-3.5-turbo

specific terms related to releases, contests, and various forms of media review

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

theless

-0.82

bably

-0.73

omever

-0.73

sic

-0.70

 disg

-0.69

 hurd

-0.68

 uphill

-0.68

fortunately

-0.67

 lowly

-0.66

maximum

-0.65

POSITIVE LOGITS

 Guide

1.10

 Dates

1.09

 Month

1.07

 Rankings

1.07

 Plans

1.06

 Plugin

1.05

 Manager

1.05

 Holder

1.03

 Pack

1.03

 Stories

1.02

Activations Density 0.309%

titles, announcements, and events

specific terms related to releases, contests, and various forms of media review

No Comments

No Known Activations