INDEX

Explanations

information related to scientific analysis or research

oai_token-act-pair · gpt-3.5-turbo

statistical data or results in scientific analyses

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

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

 Meditation

-0.77

VPN

-0.77

 vacations

-0.76

bedroom

-0.75

Cooldown

-0.75

 Skills

-0.71

 perks

-0.71

reddit

-0.70

 marches

-0.70

Theme

-0.69

POSITIVE LOGITS

 detectable

1.36

 contamin

1.33

 contamination

1.29

 concentrations

1.25

 isot

1.23

ppm

1.23

 residues

1.22

 residue

1.22

 contaminants

1.21

 specimens

1.20

Activations Density 0.332%

information related to scientific analysis or research

statistical data or results in scientific analyses

No Comments

No Known Activations