INDEX

Explanations

references to different types of cake

oai_token-act-pair · gpt-3.5-turbo

references to cake

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

nesota

-0.79

ENTION

-0.72

selves

-0.72

iveness

-0.70

ually

-0.69

audi

-0.68

 Lomb

-0.67

ostics

-0.66

 Fargo

-0.63

nen

-0.63

POSITIVE LOGITS

cakes

0.99

 cake

0.97

cake

0.94

 cakes

0.91

meal

0.90

 batter

0.83

walk

0.82

pillar

0.79

fruit

0.79

ĸļ

0.76

Activations Density 0.022%

references to different types of cake

references to cake

No Comments

No Known Activations