INDEX

Explanations

terms related to quantitative analysis or physical measurement

oai_token-act-pair · gpt-3.5-turbo

mathematical expressions or relationships involving rates and efficiencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

awks

-0.70

 Voices

-0.65

geries

-0.65

smoking

-0.65

Jere

-0.63

 Cyrus

-0.62

abad

-0.61

Boss

-0.61

izoph

-0.60

Early

-0.60

POSITIVE LOGITS

 multiplied

1.79

 divided

1.54

 minus

1.47

 squared

1.36

 calculated

1.36

/(

1.22

 normalized

1.22

 computed

1.20

(%

1.18

 multiplier

1.13

Activations Density 0.284%

terms related to quantitative analysis or physical measurement

mathematical expressions or relationships involving rates and efficiencies

No Comments

No Known Activations