INDEX

Explanations

words related to positive attributes or notable achievements

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

uttle

-0.16

elli

-0.15

taire

-0.15

hift

-0.15

awei

-0.14

ibri

-0.14

adders

-0.14

geg

-0.14

ustil

-0.13

oker

-0.13

POSITIVE LOGITS

0.16

?action

0.15

oda

0.15

atos

0.15

 sparing

0.15

iden

0.15

Bes

0.14

Bes

0.14

egen

0.14

ãĤ¾

0.14

Activations Density 0.007%

words related to positive attributes or notable achievements

No Comments

No Known Activations