INDEX

Explanations

references to different seasons of a show

oai_token-act-pair · gpt-3.5-turbo

mentions of specific seasons in a series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

pez

-0.82

izabeth

-0.82

enei

-0.79

creen

-0.79

mercial

-0.77

 sten

-0.75

wo

-0.71

inburgh

-0.71

paces

-0.69

ught

-0.68

POSITIVE LOGITS

 finale

0.84

 premiere

0.76

 Sponsor

0.76

 Tickets

0.74

 opener

0.74

 Premiere

0.71

VII

0.71

 Season

0.69

 Ticket

0.68

 Solitaire

0.68

Activations Density 0.021%

references to different seasons of a show

mentions of specific seasons in a series

No Comments

No Known Activations