INDEX

Explanations

words related to political systems and government activities

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

gur

-0.76

yz

-0.72

arta

-0.70

lain

-0.68

ingham

-0.64

ZA

-0.61

pelling

-0.60

tek

-0.59

LV

-0.59

ammy

-0.58

POSITIVE LOGITS

rils

1.22

entious

1.14

ril

0.97

 toward

0.96

entimes

0.91

 towards

0.88

 grav

0.80

erest

0.76

erer

0.75

 favour

0.75

Activations Density 0.018%

words related to political systems and government activities

No Comments

No Known Activations