INDEX

Explanations

references to a specific TV show or its characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

åİŁå§ĭ

-0.15

plier

-0.15

>{!!

-0.15

Î¸Î¼

-0.15

CodeGen

-0.15

hardware

-0.14

sock

-0.14

itness

-0.14

iosper

-0.14

Publication

-0.13

POSITIVE LOGITS

ORY

0.17

uel

0.17

ÑĦÐµÑĢ

0.16

ory

0.15

indow

0.15

 brat

0.15

 Wing

0.15

inan

0.15

æ¸¸

0.14

 Ð¤ÐµÐ´

0.14

Activations Density 0.017%

references to a specific TV show or its characters

No Comments

No Known Activations