INDEX

Explanations

names related to philosophy and historic figures or terms

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ometown

-0.90

bucks

-0.84

usercontent

-0.81

 giveaway

-0.80

daq

-0.79

NYSE

-0.78

DAQ

-0.76

 icing

-0.76

 GoPro

-0.75

 boobs

-0.74

POSITIVE LOGITS

 philosophers

1.66

 Hegel

1.55

 epist

1.55

 Aristotle

1.43

 analytic

1.43

 philosopher

1.42

 Nietzsche

1.39

 Marxism

1.37

 anthropology

1.37

 methodological

1.36

Activations Density 0.861%

names related to philosophy and historic figures or terms

No Comments

No Known Activations