INDEX

Explanations

elements related to media, news articles, and their structure

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

iras

-0.14

 Vinci

-0.14

tones

-0.14

_simps

-0.13

ridged

-0.13

otty

-0.13

usra

-0.13

.Annotations

-0.13

ichert

-0.13

ãģ¨ãĤĤ

-0.13

POSITIVE LOGITS

autoplay

0.18

 Sense

0.18

 sense

0.17

Sense

0.16

sense

0.15

 Prince

0.15

lund

0.15

GLOBAL

0.15

ä¸¤äºº

0.14

antal

0.14

Activations Density 0.112%

elements related to media, news articles, and their structure

No Comments

No Known Activations