INDEX

Explanations

words related to famous individuals

oai_token-act-pair · gpt-3.5-turbo

phrases related to automotive experiences and devices

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

zbollah

-0.66

':

-0.66

 Nato

-0.64

Newsletter

-0.62

 âĢİ

-0.61

 ÂŃ

-0.60

zinski

-0.59

],"

-0.58

Â¶

-0.58

Brexit

-0.58

POSITIVE LOGITS

 badass

0.77

 dudes

0.74

SourceFile

0.74

 HUGE

0.74

 eleph

0.70

 dude

0.70

 hilarious

0.69

 hilar

0.68

 pissed

0.68

 EVERY

0.68

Activations Density 1.884%

words related to famous individuals

phrases related to automotive experiences and devices

No Comments

No Known Activations