INDEX

Explanations

references to seasons, games, and performance statistics in sports contexts

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

ucs

-0.15

 Ð¿Ð¾Ð±

-0.15

 Rudd

-0.14

ÐµÑĦ

-0.14

krom

-0.14

essler

-0.14

ihn

-0.14

ighton

-0.14

_MT

-0.13

æ¸

-0.13

POSITIVE LOGITS

 season

0.18

ãĥ¼ãĤº

0.17

 seasons

0.16

 campaign

0.16

oin

0.14

}.{

0.14

_Msk

0.14

 Addr

0.14

season

0.14

ampaign

0.14

Activations Density 0.050%

references to seasons, games, and performance statistics in sports contexts

No Comments

No Known Activations