INDEX

Explanations

ambiguous or unrelated text

oai_token-act-pair · gpt-3.5-turbo

the phrase "what's more" or variations of it, indicating an additive or comparative emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

heid

-0.77

hyde

-0.73

xtap

-0.72

hement

-0.66

DragonMagazine

-0.66

nikov

-0.65

Ń·

-0.61

borg

-0.59

oak

-0.58

aed

-0.58

POSITIVE LOGITS

 than

1.09

ado

0.95

 stringent

0.81

 importantly

0.81

SPONSORED

0.80

 realistically

0.77

 abundantly

0.75

than

0.73

 costly

0.73

 besides

0.70

Activations Density 0.061%

ambiguous or unrelated text

the phrase "what's more" or variations of it, indicating an additive or comparative emphasis

No Comments

No Known Activations