INDEX

Explanations

references to facial features or modifications

oai_token-act-pair · gpt-3.5-turbo

references to facial features or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Breed

-0.78

 Slug

-0.76

 Seas

-0.73

 Hobby

-0.72

 Polar

-0.71

 grandfather

-0.71

UCK

-0.70

Trend

-0.69

 Salvation

-0.67

 Passage

-0.67

POSITIVE LOGITS

ilitation

1.28

ulty

1.08

fac

1.03

ilities

1.01

ilit

1.01

ility

0.98

eless

0.97

ilitating

0.91

fac

0.91

emate

0.90

Activations Density 0.012%

references to facial features or modifications

references to facial features or characteristics

No Comments

No Known Activations