INDEX

Explanations

prepositions followed by nouns for phrases related to objects and actions

oai_token-act-pair · gpt-3.5-turbo

commas and other punctuation marks in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Chiefs

-0.74

Mai

-0.72

 Nare

-0.67

oup

-0.66

 Argon

-0.66

Nom

-0.66

payer

-0.63

 Tale

-0.63

mite

-0.63

POSITIVE LOGITS

 thereby

0.83

tein

0.79

 oneself

0.78

 minimizing

0.77

 stuffing

0.75

 noticing

0.73

 preferably

0.71

 distributing

0.71

 maximizing

0.71

 optimizing

0.70

Activations Density 0.667%

prepositions followed by nouns for phrases related to objects and actions

commas and other punctuation marks in the text

No Comments

No Known Activations