INDEX

Explanations

phrases related to different versions and editions of items, such as translations, models, releases, and versions

oai_token-act-pair · gpt-3.5-turbo

references to media releases and translations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

pmwiki

-0.77

usters

-0.68

xious

-0.68

oola

-0.65

idden

-0.65

ealous

-0.63

cellaneous

-0.63

Downloadha

-0.62

ocene

-0.61

awa

-0.61

POSITIVE LOGITS

 Christensen

0.67

 abroad

0.64

 Alam

0.63

elist

0.61

 Stout

0.60

 Nath

0.57

 qualifier

0.55

 diplomat

0.55

Alm

0.55

bourg

0.55

Activations Density 0.446%

phrases related to different versions and editions of items, such as translations, models, releases, and versions

references to media releases and translations

No Comments

No Known Activations