INDEX

Explanations

adjectives and adverbs describing intensity or levels of something

oai_token-act-pair · gpt-3.5-turbo

phrases expressing strong assessments or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĥ³ãĤ¸

-0.80

querade

-0.78

 Cheong

-0.72

anwhile

-0.72

aligned

-0.69

ioned

-0.68

 defaults

-0.68

imeter

-0.66

enced

-0.64

atted

-0.64

POSITIVE LOGITS

 astounding

1.45

 remarkable

1.39

 fascinating

1.35

 impressive

1.30

 shocking

1.30

 astonishing

1.29

 priceless

1.26

 exciting

1.25

 disappointing

1.24

 amazing

1.24

Activations Density 0.279%

adjectives and adverbs describing intensity or levels of something

phrases expressing strong assessments or evaluations

No Comments

No Known Activations