INDEX

Explanations

references to restrictions or limitations

oai_token-act-pair · gpt-3.5-turbo

terms related to restrictions and their implications

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

tta

-0.85

tes

-0.84

ãĤ©

-0.81

 Briggs

-0.73

tis

-0.71

 Dirt

-0.71

da

-0.71

arist

-0.70

uca

-0.70

ISTORY

-0.70

POSITIVE LOGITS

 restricted

1.10

ricted

1.02

 unrestricted

1.01

restricted

0.92

 warr

0.88

 satell

0.85

 reserved

0.83

 lockdown

0.78

ategory

0.76

 tradem

0.75

Activations Density 0.010%

references to restrictions or limitations

terms related to restrictions and their implications

No Comments

No Known Activations