INDEX

Explanations

phrases related to scientific measurements and quantification

oai_token-act-pair · gpt-3.5-turbo

references to various types of measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

paragraph

-0.77

unct

-0.70

illard

-0.70

oci

-0.69

lined

-0.69

iled

-0.67

redo

-0.67

irl

-0.67

olic

-0.66

olia

-0.66

POSITIVE LOGITS

 measurements

1.31

 measurement

1.27

 measuring

1.17

Measure

1.11

 Measure

0.98

asuring

0.96

 measure

0.95

 glim

0.91

 Telescope

0.86

 gauge

0.86

Activations Density 0.013%

phrases related to scientific measurements and quantification

references to various types of measurements

No Comments

No Known Activations