INDEX

Explanations

words related to locations, specifically involving Washington, D.C

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

 Aval

-0.67

Aph

-0.66

 Lerner

-0.65

 actionGroup

-0.64

terday

-0.61

 Pose

-0.61

 Prelude

-0.61

 lett

-0.60

Vik

-0.59

 Situation

-0.58

POSITIVE LOGITS

ixie

1.16

isco

1.01

urga

1.00

imensional

0.98

etermination

0.97

WA

0.97

istant

0.96

ork

0.96

arts

0.94

enton

0.92

Activations Density 0.015%

words related to locations, specifically involving Washington, D.C

No Comments

No Known Activations