INDEX

Explanations

references to types of food, particularly specific dishes or ingredients

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

çĵ¶

-0.17

 cans

-0.16

 bottle

-0.16

 Bottle

-0.16

 bottles

-0.15

 peanuts

-0.15

 Candy

-0.14

 Dess

-0.14

 jars

-0.14

 faiz

-0.14

POSITIVE LOGITS

 dough

0.29

 Dough

0.23

 rolls

0.23

 turnovers

0.21

 puff

0.20

 roll

0.20

Fold

0.19

 batter

0.19

.fold

0.19

rolls

0.18

Activations Density 0.095%

references to types of food, particularly specific dishes or ingredients

No Comments

No Known Activations