INDEX

Explanations

data-related terms and information on medical conditions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

topics related to suicide and mental health statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 Rodham

-0.86

sonian

-0.82

 someday

-0.81

 Bezos

-0.78

tnc

-0.74

avored

-0.73

INTON

-0.73

soDeliveryDate

-0.71

isner

-0.71

 Lamar

-0.69

POSITIVE LOGITS

 offences

1.26

 Figures

1.03

 prosecutions

0.99

 figures

0.98

 incidents

0.97

 recorded

0.92

0.90

0.89

 statistics

0.89

0.88

Activations Density 0.539%

data-related terms and information on medical conditions

topics related to suicide and mental health statistics

No Comments

No Known Activations