INDEX

Explanations

author and publication information in academic citations

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

-0.15

 architekt

-0.15

rell

-0.14

hoe

-0.14

itals

-0.14

anz

-0.14

ATEGORIES

-0.14

ÑģÑĮÐºÐ¾

-0.14

ÙĬÙĦØ©

-0.14

Ð½Ð¾Ð¶

-0.14

POSITIVE LOGITS

trag

0.15

.struct

0.14

çĪ

0.14

cen

0.14

tin

0.14

Wer

0.13

anÄ±

0.13

âĢĮØ¨

0.13

squeeze

0.13

Activations Density 0.001%

author and publication information in academic citations

No Comments

No Known Activations