INDEX

Explanations

phrases related to locations or geographical areas

oai_token-act-pair · gpt-3.5-turbo

phrases related to significant discoveries or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

erved

-0.70

ertodd

-0.69

(?,

-0.68

 Libraries

-0.67

phrine

-0.66

 Blumenthal

-0.65

edin

-0.64

htaking

-0.64

adeon

-0.64

ariat

-0.63

POSITIVE LOGITS

 precursor

0.97

 predecessor

0.93

 successor

0.89

 notoriously

0.87

 famed

0.81

 notorious

0.81

 birthplace

0.79

 perennial

0.78

nown

0.76

 formerly

0.76

Activations Density 0.349%

phrases related to locations or geographical areas

phrases related to significant discoveries or events

No Comments

No Known Activations