INDEX

Explanations

conditional statements and questions

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

 basically

-0.15

 behalf

-0.14

vt

-0.14

ilden

-0.14

ender

-0.14

_unlock

-0.13

ç³»

-0.13

 proverb

-0.13

.djang

-0.13

andas

-0.13

POSITIVE LOGITS

 possible

0.27

 necessary

0.22

possible

0.22

Possible

0.21

 posible

0.21

 unsure

0.20

 Possible

0.20

 Ð¶Ðµ

0.20

 Necessary

0.20

_possible

0.20

Activations Density 0.092%

conditional statements and questions

No Comments

No Known Activations