INDEX

Explanations

phrases related to asserting and maintaining a firm stance on a particular topic

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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

gone

-0.82

NetMessage

-0.67

sung

-0.65

xual

-0.63

posure

-0.62

ãĥ¼ãĥ«

-0.61

ãĥŃ

-0.61

brance

-0.60

anon

-0.60

________________

-0.59

POSITIVE LOGITS

ently

0.90

ially

0.83

antly

0.77

encies

0.72

enance

0.72

 vehemently

0.72

oux

0.70

iago

0.70

 adherence

0.69

ively

0.69

Activations Density 9.488%

phrases related to asserting and maintaining a firm stance on a particular topic

No Comments

No Known Activations