INDEX

Explanations

references to the word "large" and its variations in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 smaller

-0.20

 little

-0.18

 lille

-0.17

çļĦå°ı

-0.17

 small

-0.17

 tiny

-0.17

 smallest

-0.17

 large

-0.16

 larger

-0.16

å°ı

-0.15

POSITIVE LOGITS

-scale

0.57

 scale

0.34

-format

0.32

scale

0.32

 Scale

0.28

Scale

0.28

_scale

0.26

-hearted

0.26

 SCALE

0.25

 amounts

0.24

Activations Density 0.055%

references to the word "large" and its variations in context

No Comments

No Known Activations