INDEX

Explanations

information on events involving two entities or individuals

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate simultaneous actions or occurrences related to multiple subjects or parties

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

ugu

-0.85

uez

-0.79

udic

-0.76

opian

-0.74

Ã§a

-0.73

onite

-0.71

uable

-0.71

istle

-0.71

volt

-0.70

biz

-0.70

POSITIVE LOGITS

 sexes

1.38

 sides

1.24

 halves

1.23

 genders

1.19

 parties

0.91

 sets

0.80

 thirds

0.80

 kinds

0.79

ocating

0.77

 ends

0.75

Activations Density 0.050%

information on events involving two entities or individuals

phrases that indicate simultaneous actions or occurrences related to multiple subjects or parties

No Comments

No Known Activations