INDEX

Explanations

references to scripts

oai_token-act-pair · gpt-3.5-turbo

references to scripts in various contexts, particularly in filmmaking and screenwriting

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

iscopal

-0.64

 Charity

-0.64

evil

-0.63

hips

-0.62

obby

-0.62

Ath

-0.61

hm

-0.60

 Gallup

-0.59

vic

-0.58

theless

-0.58

POSITIVE LOGITS

writers

1.19

writer

1.10

writing

1.09

ural

1.07

urally

0.98

wright

0.97

smanship

0.90

 screenplay

0.86

 synopsis

0.86

uring

0.83

Activations Density 0.009%

references to scripts

references to scripts in various contexts, particularly in filmmaking and screenwriting

No Comments

No Known Activations