INDEX

Explanations

references to the number two or multiples thereof

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

ï»¿#

-0.18

ling

-0.16

ulfilled

-0.16

adil

-0.16

ienne

-0.16

semble

-0.15

Æ°á»Ľng

-0.14

ikan

-0.14

INAL

-0.14

liness

-0.14

POSITIVE LOGITS

.infinity

0.18

ts

0.17

-double

0.17

 entend

0.16

/tr

0.16

-ended

0.15

ax

0.15

.parseDouble

0.14

opoly

0.14

cross

0.14

Activations Density 0.027%

references to the number two or multiples thereof

No Comments

No Known Activations