INDEX

Explanations

negative sentiment regarding the perception of something being forced or not of interest

oai_token-act-pair · gpt-3.5-turbo

instances of tension or conflict in narrative situations

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

 ÂŃ

-1.33

âĢĳ

-1.11

—

-1.05

Footnote

-0.87

–

-0.83

"—

-0.82

Â®,

-0.80

)—

-0.79

Enlarge

-0.79

âĢĲ

-0.79

POSITIVE LOGITS

 doesnt

1.91

 didnt

1.78

 dont

1.78

 alot

1.64

tho

1.50

 english

1.49

lol

1.47

dmg

1.39

lvl

1.36

 wont

1.31

Activations Density 1.346%

negative sentiment regarding the perception of something being forced or not of interest

instances of tension or conflict in narrative situations

No Comments

No Known Activations