INDEX

Explanations

adverbs indicating generalizations or common truths

oai_token-act-pair · gpt-3.5-turbo

phrases that express generalizations or broadly applicable statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ÄŁ

-0.70

lyn

-0.69

 Tycoon

-0.69

acity

-0.66

 Mysteries

-0.65

illion

-0.65

 Frenzy

-0.64

 Orchestra

-0.64

ilion

-0.63

poke

-0.62

POSITIVE LOGITS

 speaking

1.33

 regarded

1.00

 frowned

0.89

 accepted

0.88

speaking

0.84

 construed

0.83

 disliked

0.77

WAYS

0.75

 considered

0.75

 disappro

0.74

Activations Density 0.045%

adverbs indicating generalizations or common truths

phrases that express generalizations or broadly applicable statements

No Comments

No Known Activations