INDEX

Explanations

words related to responsibility or accountability

oai_token-act-pair · gpt-3.5-turbo

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

razil

-0.71

->

-0.61

 -------

-0.59

-->

-0.58

wcs

-0.58

 Starts

-0.57

['

-0.56

aeda

-0.56

 Emails

-0.56

achusetts

-0.55

POSITIVE LOGITS

 downright

0.83

 Lovecraft

0.77

 decidedly

0.77

 frontman

0.73

ABV

0.72

 splend

0.72

 solitude

0.70

 indul

0.70

 Flavoring

0.70

 wretched

0.69

Activations Density 1.394%

words related to responsibility or accountability

No Comments

No Known Activations

words related to responsibility or accountability

No Comments

No Known Activations