INDEX

Explanations

dishes and food-related expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pancakes

-0.16

 Panc

-0.15

_pow

-0.15

 bureaucr

-0.14

æ¾

-0.14

shake

-0.14

 dinners

-0.14

 meals

-0.14

 activeClassName

-0.14

Meal

-0.14

POSITIVE LOGITS

 hors

0.37

 finger

0.35

finger

0.31

 grazing

0.28

 bite

0.28

 Finger

0.28

nib

0.27

 spread

0.26

 spreads

0.26

 appet

0.25

Activations Density 0.108%

dishes and food-related expressions

No Comments

No Known Activations

dishes and food-related expressions

No Comments

No Known Activations