INDEX

Explanations

mentions of security breaches or violations

oai_token-act-pair · gpt-3.5-turbo

references to breaches or violations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

uana

-0.86

zl

-0.77

ICA

-0.69

arij

-0.67

lee

-0.65

minist

-0.65

onga

-0.65

 Towns

-0.65

 stag

-0.64

puff

-0.62

POSITIVE LOGITS

 breaches

0.87

 breach

0.81

terness

0.80

ingly

0.76

hole

0.76

hold

0.75

holes

0.74

 perpetrated

0.70

ionage

0.70

taking

0.70

Activations Density 0.038%

mentions of security breaches or violations

references to breaches or violations

No Comments

No Known Activations