INDEX

Explanations

mentions of miscellaneous categories and classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ÐµÑĤÑĮ

-0.18

-0.16

Ð½Ð¸ÑĨ

-0.15

GLOBALS

-0.15

ÐĽÐ¬

-0.14

sted

-0.14

ice

-0.14

sst

-0.14

GetInstance

-0.13

iska

-0.13

POSITIVE LOGITS

/misc

0.26

ellaneous

0.25

misc

0.23

Misc

0.22

 misc

0.21

 miscellaneous

0.20

 general

0.19

/general

0.19

 Misc

0.18

-purpose

0.16

Activations Density 0.101%

mentions of miscellaneous categories and classifications

No Comments

No Known Activations