INDEX

Explanations

numeric data that indicates quantity or count

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

antas

-0.14

sson

-0.14

ssize

-0.14

coords

-0.13

 Alternate

-0.13

 DbSet

-0.13

 _↵↵

-0.13

Å

-0.13

 ÙĨÙĤ

-0.13

ERG

-0.13

POSITIVE LOGITS

enze

0.18

baugh

0.15

ceptar

0.15

/detail

0.14

.motion

0.14

lich

0.14

Ð¾Ð»Ð¾Ñģ

0.14

qd

0.13

 wrench

0.13

unl

0.13

Activations Density 0.000%

numeric data that indicates quantity or count

No Comments

No Known Activations