INDEX

Explanations

phrases related to email alerts and news updates

oai_token-act-pair · gpt-3.5-turbo

references to news alerts and email notifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

itar

-1.00

stood

-0.84

own

-0.83

ãĥĵ

-0.81

folk

-0.79

fig

-0.78

shr

-0.77

jury

-0.76

ded

-0.75

inav

-0.73

POSITIVE LOGITS

 Emails

1.36

 Transcript

0.97

 Timeline

0.90

VERTISEMENT

0.89

querque

0.83

ļéĨĴ

0.78

 Loading

0.73

 transcripts

0.72

 tiss

0.72

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

0.72

Activations Density 0.007%

phrases related to email alerts and news updates

references to news alerts and email notifications

No Comments

No Known Activations