INDEX

Explanations

mentions of various unrelated topics

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

ÄŁ

-0.64

letes

-0.55

lyn

-0.54

Kyl

-0.54

ilion

-0.54

 Manz

-0.53

Ced

-0.53

 Ferry

-0.52

hao

-0.52

 Seat

-0.51

POSITIVE LOGITS

 speaking

0.81

 regarded

0.77

 frowned

0.68

 accepted

0.65

 appreciated

0.63

WAYS

0.62

ensical

0.61

 categorized

0.60

 disliked

0.59

 considered

0.58

Activations Density 9.208%

mentions of various unrelated topics

No Comments

No Known Activations