INDEX

Explanations

the word "and" and its variations in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 even

-0.17

InOut

-0.15

 EVEN

-0.15

aille

-0.14

este

-0.14

ince

-0.14

eliness

-0.13

.glide

-0.13

Verse

-0.13

lein

-0.13

POSITIVE LOGITS

/or

0.24

rew

0.24

 zwar

0.23

rea

0.20

REW

0.18

reas

0.17

ystate

0.16

hra

0.16

ROID

0.16

erval

0.16

Activations Density 0.171%

the word "and" and its variations in a document

No Comments

No Known Activations