INDEX

Explanations

mentions of specific individuals, especially in contexts of significance or notoriety

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

pher

-0.66

 Opera

-0.60

 Denmark

-0.60

Tat

-0.59

 Malays

-0.57

 exile

-0.57

 Integrity

-0.56

 Arabs

-0.56

 Thieves

-0.55

 rough

-0.55

POSITIVE LOGITS

acks

1.03

idy

0.99

essions

0.93

essee

0.91

enger

0.89

essing

0.84

imus

0.82

ips

0.81

essed

0.80

umbnails

0.79

Activations Density 0.177%

mentions of specific individuals, especially in contexts of significance or notoriety

No Comments

No Known Activations