INDEX

Explanations

words related to modifications or alterations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lauder

-0.66

izabeth

-0.63

 Mata

-0.63

 Bulldogs

-0.63

 leukemia

-0.61

Lay

-0.61

 Treasurer

-0.61

 Bucc

-0.60

 Lilly

-0.60

lihood

-0.60

POSITIVE LOGITS

elled

1.43

ded

1.40

ding

1.38

ulo

1.35

icum

1.34

ulators

1.33

ulations

1.33

ifiable

1.30

ulus

1.27

ulated

1.27

Activations Density 0.032%

words related to modifications or alterations

No Comments

No Known Activations