INDEX

Explanations

phrases related to events and places happening across a wide geographic area

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pmwiki

-0.77

_-

-0.67

 guiActive

-0.64

BIL

-0.61

 chau

-0.61

CVE

-0.61

Charges

-0.60

DUI

-0.58

 alarms

-0.58

isons

-0.58

POSITIVE LOGITS

 globe

1.40

 country

1.23

 continents

1.13

 world

1.10

 continent

1.09

country

1.01

 nation

0.99

 periphery

0.93

 planet

0.92

astern

0.92

Activations Density 2.179%

phrases related to events and places happening across a wide geographic area

No Comments

No Known Activations