INDEX

Explanations

phrases related to podcasts

oai_token-act-pair · gpt-3.5-turbo

references to podcasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ju

-0.72

metics

-0.69

arial

-0.69

yer

-0.65

arde

-0.64

 curv

-0.64

axy

-0.63

forces

-0.63

illard

-0.61

come

-0.61

POSITIVE LOGITS

 listeners

1.01

odcast

0.99

 podcasts

0.94

 podcast

0.89

 transcript

0.89

podcast

0.89

 listener

0.87

estern

0.83

Episode

0.83

 episode

0.82

Activations Density 0.041%

phrases related to podcasts

references to podcasts

No Comments

No Known Activations