INDEX

Explanations

technical symbols or elements in mathematical or programming contexts

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

al

-0.16

 inclusive

-0.15

rieve

-0.14

 vict

-0.14

 Moment

-0.13

 imports

-0.13

uzu

-0.13

ÃŃr

-0.13

bbie

-0.13

ACHED

-0.13

POSITIVE LOGITS

ä»ķ

0.19

 Lust

0.15

cream

0.15

 dzie

0.14

alah

0.14

ìĭŃ

0.14

ÏĢÏĮ

0.14

_reserved

0.14

ummings

0.14

éª

0.14

Activations Density 0.000%

technical symbols or elements in mathematical or programming contexts

No Comments

No Known Activations