INDEX

Explanations

verbs in the infinitive form that describe actions or instructions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TION

-1.07

listed

-1.00

hair

-1.00

bub

-0.96

 resemb

-0.91

 stru

-0.91

tek

-0.91

adv

-0.91

iana

-0.90

 awoke

-0.90

POSITIVE LOGITS

 minimize

1.34

 eliminate

1.31

 ensure

1.27

asted

1.24

 avoid

1.24

 accommodate

1.24

ilet

1.23

pless

1.22

 maximize

1.19

 save

1.17

Activations Density 0.769%

verbs in the infinitive form that describe actions or instructions

No Comments

No Known Activations