INDEX

Explanations

mentions of speakers or presenters in various events or contexts

oai_token-act-pair · gpt-3.5-turbo

references to speakers and presentations at events or workshops

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 ILCS

-0.69

agall

-0.67

 porous

-0.67

 thicker

-0.64

 surg

-0.61

anking

-0.61

 everlasting

-0.60

vernment

-0.59

 destroyer

-0.59

 batter

-0.58

POSITIVE LOGITS

 presentations

1.39

 attendees

1.39

 keynote

1.33

 lectures

1.25

 lecture

1.14

 seminars

1.13

inar

1.10

 seminar

1.08

 Lect

1.07

posium

1.06

Activations Density 0.283%

mentions of speakers or presenters in various events or contexts

references to speakers and presentations at events or workshops

No Comments

No Known Activations