INDEX

Explanations

terms related to the word "rum"

oai_token-act-pair · gpt-3.5-turbo

references to "rum," particularly in various contexts

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

 prevail

-0.68

 clean

-0.67

 inclusive

-0.65

 crawl

-0.65

 privileges

-0.64

 loading

-0.64

 safe

-0.63

 save

-0.63

;;;;;;;;;;;;

-0.63

 prompt

-0.62

POSITIVE LOGITS

rum

4.93

rums

2.99

Rum

1.55

ram

1.30

Rum

1.28

ral

1.26

ril

1.23

rick

1.23

rim

1.21

rum

1.16

Activations Density 0.014%

terms related to the word "rum"

references to "rum," particularly in various contexts

No Comments

No Known Activations