INDEX

Explanations

words related to meaning, merit, and worth

oai_token-act-pair · gpt-3.5-turbo

concepts related to meaning, worth, and merit

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Aren

-0.65

 Rove

-0.64

Sun

-0.63

 Franch

-0.62

 Kham

-0.61

 Tanz

-0.60

 Violet

-0.60

 Colossus

-0.59

Bed

-0.59

 Wake

-0.59

POSITIVE LOGITS

fully

2.06

lessness

1.95

ful

1.88

fulness

1.88

lessly

1.86

less

1.79

full

1.51

iness

1.34

FUL

1.32

ific

1.31

Activations Density 0.147%

words related to meaning, merit, and worth

concepts related to meaning, worth, and merit

No Comments

No Known Activations