INDEX

Explanations

mentions of villages

oai_token-act-pair · gpt-3.5-turbo

mentions of villages

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

Negative Logits

 Astro

-0.74

 Stern

-0.67

immer

-0.66

BB

-0.63

 Astros

-0.63

AUD

-0.63

 Eminem

-0.62

trump

-0.62

 cancel

-0.62

BRE

-0.61

POSITIVE LOGITS

 villages

3.23

 village

2.97

 villagers

2.65

 Village

1.88

 vill

1.84

Vill

1.79

 countryside

1.66

 towns

1.63

 peasants

1.58

 township

1.51

Activations Density 0.021%

mentions of villages

mentions of villages

No Known Activations