INDEX

Explanations

references to podcasts, interviews, and media productions

oai_token-act-pair · gpt-3.5-turbo

references to podcast episodes and discussions

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

igslist

-0.72

brance

-0.70

ativity

-0.70

 utilitarian

-0.69

assault

-0.68

otrop

-0.67

oufl

-0.67

owment

-0.67

mop

-0.67

 altru

-0.67

POSITIVE LOGITS

 listener

1.30

 Topics

1.28

Episode

1.23

 recap

1.21

 Recap

1.17

 Episode

1.16

 Podcast

1.14

 Tune

1.13

 listeners

1.12

Topics

1.08

Activations Density 0.541%

references to podcasts, interviews, and media productions

references to podcast episodes and discussions

No Comments

No Known Activations