INDEX

Explanations

phrases related to publications or works such as imprints, memoirs, and battlegrounds

oai_token-act-pair · gpt-3.5-turbo

terms related to personal relationships and significant events

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

astern

-0.69

ULTS

-0.66

abases

-0.62

 Roses

-0.62

IPP

-0.58

Downloadha

-0.58

OPLE

-0.57

flies

-0.56

ploma

-0.56

ELF

-0.56

POSITIVE LOGITS

2.72

sburg

1.45

sie

1.42

ski

1.40

sb

1.36

sin

1.29

sg

1.26

sand

1.23

sf

1.23

sburgh

1.16

Activations Density 0.286%

phrases related to publications or works such as imprints, memoirs, and battlegrounds

terms related to personal relationships and significant events

No Comments

No Known Activations