INDEX

Explanations

references to entertainment industry publications

oai_token-act-pair · gpt-3.5-turbo

references to film industry publications and their reporting

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

alist

-0.82

quer

-0.76

İĭ

-0.75

carry

-0.74

isl

-0.72

chest

-0.69

plan

-0.68

train

-0.68

raf

-0.68

alias

-0.67

POSITIVE LOGITS

 Reporter

0.92

 RELEASE

0.85

 Variety

0.83

 Deadline

0.79

 Awards

0.75

 Release

0.71

 Editors

0.71

 synopsis

0.70

 RELE

0.69

 Wrap

0.69

Activations Density 0.027%

references to entertainment industry publications

references to film industry publications and their reporting

No Comments

No Known Activations