INDEX

Explanations

locations or places with a specific word in their name

oai_token-act-pair · gpt-3.5-turbo

mentions of various "stations."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ily

-0.82

Iv

-0.76

orial

-0.72

ric

-0.72

 unic

-0.67

 inherit

-0.66

 spelling

-0.65

 inher

-0.64

ILY

-0.63

dh

-0.63

POSITIVE LOGITS

 Station

3.89

 station

2.55

Station

2.44

station

2.33

 stations

2.26

 Terminal

1.33

 Airport

1.28

 Depot

1.26

 Junction

1.12

 Railway

1.09

Activations Density 0.008%

locations or places with a specific word in their name

mentions of various "stations."

No Comments

No Known Activations