INDEX

Explanations

items that are part of a series or installment

oai_token-act-pair · gpt-3.5-turbo

references to sequels or series of works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 guessed

-0.71

Sport

-0.69

 danced

-0.68

bos

-0.68

 likeness

-0.67

Zup

-0.65

 Provided

-0.64

 faire

-0.63

pter

-0.62

 spiders

-0.62

POSITIVE LOGITS

series

0.86

 epis

0.74

 series

0.74

 ongoing

0.72

 intensive

0.69

 saga

0.69

arte

0.65

 installment

0.65

 installments

0.65

 trilogy

0.64

Activations Density 0.221%

items that are part of a series or installment

references to sequels or series of works

No Comments

No Known Activations