INDEX

Explanations

numerical and place-related information

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

roken

-0.18

otine

-0.18

lun

-0.16

otch

-0.15

illac

-0.15

iola

-0.15

graf

-0.14

untime

-0.14

apot

-0.14

anton

-0.14

POSITIVE LOGITS

glo

0.17

bou

0.15

_argument

0.14

IVO

0.14

à¸¸à¸£

0.14

perm

0.13

Opp

0.13

Address

0.13

Unnamed

0.13

ombat

0.13

Activations Density 0.092%

numerical and place-related information

No Comments

No Known Activations