INDEX

Explanations

proper nouns related to the color "Purple"

oai_token-act-pair · gpt-3.5-turbo

occurrences and references to the color purple

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

bsp

-0.82

-0.75

yers

-0.74

elong

-0.72

ITNESS

-0.70

areth

-0.70

LER

-0.69

onal

-0.68

igmat

-0.68

ernandez

-0.68

POSITIVE LOGITS

 velvet

0.92

tail

0.86

 violet

0.85

hue

0.80

dye

0.78

grass

0.78

fruit

0.76

ink

0.76

 purple

0.75

 Purple

0.75

Activations Density 0.010%

proper nouns related to the color "Purple"

occurrences and references to the color purple

No Comments

No Known Activations