INDEX

Explanations

references to spa-related terminology and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sail

-0.16

storm

-0.16

.define

-0.16

 storm

-0.15

 Wake

-0.15

orage

-0.14

 Ø¬Ø§Ùħ

-0.14

çĩĥ

-0.14

 sails

-0.14

 Storm

-0.14

POSITIVE LOGITS

spa

0.37

Spa

0.33

 springs

0.33

 Springs

0.30

spa

0.28

 baths

0.27

 bath

0.27

 Baths

0.25

hot

0.24

 spas

0.24

Activations Density 0.040%

references to spa-related terminology and locations

No Comments

No Known Activations