INDEX

Explanations

phrases related to sending messages or signals in various contexts

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

theless

-0.78

 includ

-0.66

bid

-0.66

È

-0.64

 submitted

-0.63

ournal

-0.63

iar

-0.63

 editions

-0.62

Submit

-0.62

bery

-0.60

POSITIVE LOGITS

 signal

1.10

 ripple

1.05

 signals

1.03

 message

1.01

 shock

0.98

message

0.87

 reverber

0.82

 vibrations

0.80

shock

0.79

ipples

0.77

Activations Density 0.042%

phrases related to sending messages or signals in various contexts

No Comments

No Known Activations