INDEX

Explanations

references to specific years or events associated with dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

zp

-0.17

zac

-0.15

ÐµÐ¼

-0.15

abei

-0.15

isArray

-0.15

okane

-0.15

zas

-0.14

ufen

-0.14

 Grammar

-0.14

ipur

-0.14

POSITIVE LOGITS

iction

0.21

ano

0.20

ually

0.20

udit

0.20

igne

0.20

ite

0.19

ure

0.19

Ã»

0.19

antes

0.18

ore

0.17

Activations Density 0.007%

references to specific years or events associated with dates

No Comments

No Known Activations