INDEX

Explanations

phrases indicating future actions or outcomes

oai_token-act-pair · gpt-3.5-turbo

expressions of future possibilities or predictions

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

Â¶

-0.63

assin

-0.60

ilaterally

-0.59

 sket

-0.58

culus

-0.56

 Dude

-0.56

 Advertisement

-0.56

 specificity

-0.55

ELF

-0.54

 disadvant

-0.52

POSITIVE LOGITS

 continue

1.50

continue

1.27

 hereafter

1.18

 continuing

1.08

 continued

1.04

 soon

1.04

 resume

1.03

 forever

0.99

 continues

0.99

 remain

0.97

Activations Density 0.257%

phrases indicating future actions or outcomes

expressions of future possibilities or predictions

No Comments

No Known Activations