INDEX

Explanations

words related to trailers or previews for movies or shows

oai_token-act-pair · gpt-3.5-turbo

references to "trailer" or "traffic" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 otherwise

-0.72

lain

-0.66

nah

-0.59

 numbering

-0.58

THING

-0.58

 numbered

-0.57

lists

-0.56

 cracking

-0.56

 Brook

-0.54

 Galactic

-0.54

POSITIVE LOGITS

ffic

1.40

pped

1.37

umat

1.29

pping

1.28

ppings

1.25

umatic

1.21

pez

1.21

ilers

1.18

iler

1.14

eger

1.13

Activations Density 0.020%

words related to trailers or previews for movies or shows

references to "trailer" or "traffic" in various contexts

No Comments

No Known Activations