INDEX

Explanations

mentions of specific events and organizations

oai_token-act-pair · gpt-3.5-turbo

content related to specific events or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

[_

-0.52

Sov

-0.47

entimes

-0.45

ographically

-0.45

odynam

-0.45

imately

-0.44

olicy

-0.43

endi

-0.42

ographical

-0.42

ittees

-0.41

POSITIVE LOGITS

 UNITED

0.44

MEN

0.43

 Quit

0.38

GN

0.36

MID

0.35

ARM

0.35

CFR

0.35

 Gorge

0.35

CST

0.34

 formerly

0.34

Activations Density 5.147%

mentions of specific events and organizations

content related to specific events or activities

No Comments

No Known Activations