INDEX

Explanations

structured elements in writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

avir

-0.16

avian

-0.16

urum

-0.16

ãĤ«ãĥ¼

-0.16

adel

-0.15

 appro

-0.15

chw

-0.14

ÌĢ

-0.14

 Dumpster

-0.13

icz

-0.13

POSITIVE LOGITS

0.17

 INTERRU

0.14

Å¡t

0.14

olia

0.14

uddle

0.14

amed

0.13

ÑĪÐ¸Ð±

0.13

ÑģÑĤÐ²Ð°

0.13

aggi

0.13

Activations Density 0.265%

structured elements in writing

No Comments

No Known Activations