INDEX

Explanations

locations within a city

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

henko

-0.75

 vectors

-0.73

merga

-0.73

TPP

-0.71

Animation

-0.69

gans

-0.69

rils

-0.68

oeuv

-0.65

wana

-0.65

/(

-0.65

POSITIVE LOGITS

 downtown

1.24

 Downtown

1.09

 lieu

1.05

 suburban

1.04

 Melbourne

1.04

 Arlington

1.03

 Dublin

1.03

 Notting

1.02

 Manhattan

1.01

 southwest

1.01

Activations Density 0.143%

locations within a city

No Comments

No Known Activations