INDEX

Explanations

verbs that indicate duration or effort

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.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

 myster

-0.78

ilings

-0.76

 Kamp

-0.75

 repression

-0.75

 marqu

-0.73

 burner

-0.73

 Phant

-0.72

 Palest

-0.71

 Britann

-0.71

ieg

-0.71

POSITIVE LOGITS

aways

1.22

overs

1.18

awed

1.01

 advantage

0.99

FINE

0.94

 Mehran

0.88

 heed

0.87

autions

0.86

ngth

0.86

ave

0.85

Activations Density 0.134%

verbs that indicate duration or effort

No Comments

No Known Activations