INDEX

Explanations

references to technical details such as model names, specific locations, and codes

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Hicks

-0.95

 Hatt

-0.85

ipp

-0.82

 brace

-0.80

 bracelet

-0.75

 Hamas

-0.75

ank

-0.75

-0.74

Brook

-0.74

 Huang

-0.74

POSITIVE LOGITS

1.62

Vs

1.45

VA

1.42

1.37

VD

1.37

VC

1.33

Vit

1.33

1.30

VG

1.30

VP

1.27

Activations Density 0.535%

references to technical details such as model names, specific locations, and codes

No Comments

No Known Activations