INDEX

Explanations

references to news reporting and journalistic sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

áº§m

-0.15

throp

-0.15

à¥Ĥà¤¤

-0.14

à¸£à¸°

-0.14

oÄŁ

-0.14

 deactivated

-0.14

-0.13

aea

-0.13

oog

-0.13

POSITIVE LOGITS

urst

0.15

oucher

0.15

CEE

0.14

ç¥Ń

0.14

AVOR

0.14

dpi

0.14

clr

0.14

 envi

0.14

ÑĥÐºÐ¸

0.14

Fade

0.14

Activations Density 0.004%

references to news reporting and journalistic sources

No Comments

No Known Activations