INDEX

Explanations

references to specific racing events, teams, and personalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

sdale

-0.16

oot

-0.15

wagon

-0.14

IME

-0.14

æĪ¶

-0.14

ritz

-0.14

etas

-0.13

 personel

-0.13

à¸Ĺà¸Ńà¸ĩ

-0.13

Ð¶ÐµÐ½ÑĮ

-0.13

POSITIVE LOGITS

anford

0.16

aru

0.16

 race

0.15

aret

0.13

 vant

0.13

 technological

0.13

arov

0.13

_PARAMETERS

0.13

 Valk

0.13

 Ø§ÙĦÙĤØ±

0.13

Activations Density 0.011%

references to specific racing events, teams, and personalities

No Comments

No Known Activations