INDEX

Explanations

comparisons using the word "than."

oai_token-act-pair · gpt-3.5-turbo

comparative expressions emphasizing "more than" or "less than" comparisons

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

umbn

-0.83

estern

-0.78

illary

-0.76

onomy

-0.70

auri

-0.70

esm

-0.67

ero

-0.65

uto

-0.65

ango

-0.64

eni

-0.63

POSITIVE LOGITS

 anything

1.24

any

1.05

 anyone

0.94

 anybody

0.93

 necessarily

0.85

 ever

0.81

 usual

0.81

 anywhere

0.80

 anymore

0.69

 vice

0.69

Activations Density 0.083%

comparisons using the word "than."

comparative expressions emphasizing "more than" or "less than" comparisons

No Comments

No Known Activations