INDEX

Explanations

mentions of physical or emotional torment

oai_token-act-pair · gpt-3.5-turbo

instances of the word "tor" and related forms in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sheets

-0.73

compliance

-0.68

enforcement

-0.68

Keefe

-0.65

WER

-0.64

ortment

-0.64

 Dakota

-0.63

endif

-0.63

)=(

-0.62

enegger

-0.62

POSITIVE LOGITS

onto

1.14

rent

0.90

mented

0.89

tor

0.86

rance

0.84

atra

0.82

ques

0.79

seys

0.76

roying

0.76

chest

0.75

Activations Density 0.010%

mentions of physical or emotional torment

instances of the word "tor" and related forms in various contexts

No Comments

No Known Activations