INDEX

Explanations

phrases related to written scripts

oai_token-act-pair · gpt-3.5-turbo

mentions of scripts or screenplays

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

obby

-0.78

theless

-0.75

iscopal

-0.69

evil

-0.65

ternity

-0.63

Ath

-0.62

 Charity

-0.60

ktop

-0.60

hm

-0.59

hate

-0.59

POSITIVE LOGITS

writers

1.08

writer

1.04

urally

1.03

ural

1.03

writing

0.95

 script

0.88

uring

0.87

 screenplay

0.86

 synopsis

0.86

wright

0.86

Activations Density 0.009%

phrases related to written scripts

mentions of scripts or screenplays

No Comments

No Known Activations