INDEX

Explanations

specific keyword "find" in sentences

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate locating or accessing information

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

assisted

-0.75

istry

-0.73

aired

-0.68

 bailed

-0.65

pir

-0.63

wake

-0.62

 Mandatory

-0.62

 coerc

-0.62

pires

-0.61

oiler

-0.60

POSITIVE LOGITS

çīĪ

0.82

 plenty

0.81

ãĤĮ

0.80

DragonMagazine

0.78

ById

0.78

ãħĭ

0.75

ļé

0.75

TING

0.75

ä½¿

0.74

IU

0.72

Activations Density 0.049%

specific keyword "find" in sentences

phrases that indicate locating or accessing information

No Comments

No Known Activations