INDEX

Explanations

terms and phrases related to identity and classification

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

Ø¯Ø§ÙĨÙĦÙĪØ¯

-0.16

 Holmes

-0.15

obra

-0.14

 Statue

-0.14

LEX

-0.14

 neod

-0.13

isher

-0.13

ìĦł

-0.13

olv

-0.13

 Fres

-0.13

POSITIVE LOGITS

egov

0.15

¥

0.15

avel

0.15

.decorators

0.14

KeyType

0.14

.timedelta

0.14

gado

0.14

hana

0.13

YTE

0.13

 sadd

0.13

Activations Density 0.086%

terms and phrases related to identity and classification

No Comments

No Known Activations