INDEX

Explanations

references to possessive pronouns like 'your'

oai_token-act-pair · gpt-3.5-turbo

references to the reader or audience

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Chrys

-0.71

laus

-0.70

acca

-0.68

essee

-0.67

inctions

-0.66

 Slay

-0.66

 Trilogy

-0.63

forth

-0.62

-0.60

hell

-0.59

POSITIVE LOGITS

 device

0.91

 browser

0.86

 devices

0.72

 desktop

0.71

 unsupported

0.70

 firewall

0.69

 browsers

0.67

own

0.67

 transmitter

0.66

device

0.66

Activations Density 0.014%

references to possessive pronouns like 'your'

references to the reader or audience

No Comments

No Known Activations