INDEX

Explanations

terms related to control and management

oai_token-act-pair · gpt-3.5-turbo

terms related to management and control within various contexts

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

ĪĴ

-0.88

BuyableInstoreAndOnline

-0.69

pired

-0.68

qualified

-0.63

pires

-0.61

¿½

-0.61

Uncommon

-0.60

 Gleaming

-0.60

nces

-0.59

 honorable

-0.59

POSITIVE LOGITS

 accordingly

1.24

 differently

1.21

 movements

1.00

 through

0.95

 appropriately

0.95

 independently

0.90

 responsibly

0.90

via

0.90

 accurately

0.90

 efficiently

0.88

Activations Density 0.455%

terms related to control and management

terms related to management and control within various contexts

No Comments

No Known Activations