INDEX

Explanations

phrases related to conspiracies or evil schemes

oai_token-act-pair · gpt-3.5-turbo

references to "plot" 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

IDA

-0.71

 salts

-0.68

agles

-0.66

Scot

-0.64

Downloadha

-0.63

 Splash

-0.61

Rio

-0.61

Sales

-0.60

Occ

-0.60

angelo

-0.60

POSITIVE LOGITS

ters

0.99

 Plot

0.87

 twists

0.86

 Twist

0.82

line

0.82

 hatched

0.81

ter

0.81

 plot

0.79

 plotting

0.78

zag

0.78

Activations Density 0.031%

phrases related to conspiracies or evil schemes

references to "plot" in various contexts

No Comments

No Known Activations