INDEX

Explanations

instances of exits and exit-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Ð¾Ð»Ð¾Ð¶

-0.15

icket

-0.14

jer

-0.14

á»ĳng

-0.13

Ð°ÑĢÑĩ

-0.13

ÑģÑĤÐ°Ð½

-0.13

udev

-0.13

 Ð¾ÐºÑĢÑĥÐ¶

-0.13

ä¸ģ

-0.13

uka

-0.13

POSITIVE LOGITS

keh

0.18

 exits

0.17

0.16

 exit

0.16

anship

0.15

Ãłm

0.15

.Depth

0.15

 Exit

0.14

.cz

0.14

exit

0.14

Activations Density 0.010%

instances of exits and exit-related terminology

No Comments

No Known Activations