INDEX

Explanations

mentions of legal proceedings or actions related to law firms

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

theless

-1.10

FTWARE

-0.94

ktop

-0.92

vous

-0.90

enegger

-0.89

 Cuomo

-0.88

beard

-0.85

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.84

å£«

-0.83

BuyableInstoreAndOnline

-0.83

POSITIVE LOGITS

itudes

1.60

ainment

1.57

ention

1.57

achment

1.56

itude

1.55

ractive

1.51

ributed

1.50

ension

1.50

ributes

1.50

ribute

1.44

Activations Density 5.249%

mentions of legal proceedings or actions related to law firms

No Comments

No Known Activations