INDEX

Explanations

references to significant locations or entities related to history and culture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ureau

-0.14

rogram

-0.14

istrovstvÃŃ

-0.14

LBL

-0.13

OffsetTable

-0.13

éĥ¨åĪĨ

-0.13

 GOODMAN

-0.13

åªĴ

-0.13

rometer

-0.13

quirer

-0.13

POSITIVE LOGITS

ankind

0.25

oland

0.22

 itself

0.19

istan

0.18

ERICA

0.17

frica

0.17

-town

0.17

ussia

0.17

ustria

0.16

zelf

0.16

Activations Density 0.284%

references to significant locations or entities related to history and culture

No Comments

No Known Activations