INDEX

Explanations

phrases that indicate reports or findings in the media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ime

-0.16

vel

-0.15

agement

-0.14

æĮ¯ãĤĬ

-0.14

ching

-0.14

ellig

-0.13

 Boulevard

-0.13

bes

-0.13

CHA

-0.13

æŁĦ

-0.13

POSITIVE LOGITS

leÅ¾it

0.16

iros

0.16

ãĥ³ãĥĦ

0.16

rawer

0.15

peria

0.15

raud

0.15

rani

0.15

RIX

0.14

/Dk

0.14

ristol

0.14

Activations Density 0.175%

phrases that indicate reports or findings in the media

No Comments

No Known Activations