INDEX

Explanations

references to New Zealand

oai_token-act-pair · gpt-3.5-turbo

references to New Zealand and its geographical locations

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

itbart

-0.79

Magikarp

-0.78

abet

-0.74

igue

-0.73

mentation

-0.72

rooms

-0.71

RAW

-0.71

baugh

-0.70

Ô

-0.69

due

-0.69

POSITIVE LOGITS

 Zealand

1.38

NZ

1.33

NZ

1.12

 Auckland

1.12

Kiw

1.00

uckland

0.88

XT

0.87

 Labour

0.86

 Zeal

0.84

 Wellington

0.81

Activations Density 0.007%

references to New Zealand

references to New Zealand and its geographical locations

No Comments

No Known Activations