INDEX

Explanations

phrases related to top rankings or positions

oai_token-act-pair · gpt-3.5-turbo

phrases related to high rankings or top-tier quality

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

Dia

-0.73

 Sawyer

-0.66

Cah

-0.64

 spells

-0.64

 Wilson

-0.62

 Cavern

-0.62

Syl

-0.62

SLI

-0.61

Cox

-0.61

 Vaughn

-0.61

POSITIVE LOGITS

tier

1.56

ranked

1.47

level

1.47

ranking

1.46

performing

1.35

priority

1.30

rated

1.28

scoring

1.26

selling

1.26

five

1.23

Activations Density 0.019%

phrases related to top rankings or positions

phrases related to high rankings or top-tier quality

No Comments

No Known Activations