INDEX

Explanations

accusations or blame directed towards individuals or entities

oai_token-act-pair · gpt-3.5-turbo

instances of blame and accusations directed at individuals, groups, or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æĸ¹

-0.78

redo

-0.69

OSH

-0.69

erves

-0.65

daq

-0.65

 largeDownload

-0.64

ibaba

-0.63

 welcomes

-0.63

iosis

-0.61

areth

-0.61

POSITIVE LOGITS

 inaction

0.88

à¨

0.80

ãĥ³ãĤ¸

0.79

 faulty

0.68

 coward

0.67

 negligence

0.66

ackers

0.65

 traffickers

0.65

 accuser

0.64

 cowardly

0.64

Activations Density 0.337%

accusations or blame directed towards individuals or entities

instances of blame and accusations directed at individuals, groups, or entities

No Comments

No Known Activations