INDEX

Explanations

words related to quantities and measurements

oai_token-act-pair · gpt-3.5-turbo

terms related to inventories or stock levels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 apost

-0.65

 opponent

-0.62

 aggro

-0.59

 Visa

-0.57

 respectfully

-0.57

 vocal

-0.57

sir

-0.57

 stride

-0.56

 positioned

-0.56

 visa

-0.56

POSITIVE LOGITS

ories

5.23

ory

2.77

ORY

2.65

orie

2.17

orians

1.58

orical

1.52

oire

1.46

orial

1.46

oret

1.36

oric

1.30

Activations Density 0.009%

words related to quantities and measurements

terms related to inventories or stock levels

No Comments

No Known Activations