INDEX

Explanations

phrases indicating conditions or scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

enerator

-0.15

 Aires

-0.14

roman

-0.14

 sice

-0.14

umd

-0.13

indo

-0.13

berger

-0.13

atsu

-0.13

ãĥ³ãĥĩãĤ£

-0.13

 rencontre

-0.13

POSITIVE LOGITS

arken

0.16

adaÅŁ

0.15

LTR

0.15

ebek

0.15

Ïį

0.14

æĿ¥è¯´

0.14

edian

0.14

aight

0.14

Î¼Î²

0.14

ypi

0.14

Activations Density 0.206%

phrases indicating conditions or scenarios

No Comments

No Known Activations