INDEX

Explanations

geographical locations, specifically those containing the word "Sur."

oai_token-act-pair · gpt-3.5-turbo

references to a specific location or area named "Sur."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

naires

-0.94

 Archdemon

-0.78

INESS

-0.75

EMENT

-0.67

isle

-0.64

 INFO

-0.64

Constructed

-0.63

 Goddard

-0.62

 Eternity

-0.62

umbn

-0.61

POSITIVE LOGITS

geon

1.37

geons

1.13

render

1.05

rounded

1.01

fer

0.96

round

0.88

fing

0.87

rito

0.85

amba

0.85

ports

0.84

Activations Density 0.008%

geographical locations, specifically those containing the word "Sur."

references to a specific location or area named "Sur."

No Comments

No Known Activations