INDEX

Explanations

the word "you"

oai_token-act-pair · gpt-3.5-turbo

references to the second person pronoun "you."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 airs

-0.67

acular

-0.65

math

-0.64

 Sabha

-0.63

ice

-0.62

 Reconstruction

-0.62

 Richmond

-0.62

 Associated

-0.62

ensable

-0.61

 Samoa

-0.61

POSITIVE LOGITS

're

1.52

've

1.31

'll

1.23

 guys

1.15

'd

1.04

tub

1.01

 know

1.00

 want

0.92

 guessed

0.92

hei

0.92

Activations Density 0.310%

the word "you"

references to the second person pronoun "you."

No Comments

No Known Activations