INDEX

Explanations

references to structural components and their positions within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

mina

-0.15

ãĥ³ãĥĨãĤ£

-0.15

ccoli

-0.14

locking

-0.14

staking

-0.14

ycl

-0.14

elon

-0.14

.palette

-0.14

IGHLIGHT

-0.14

avan

-0.13

POSITIVE LOGITS

 bottom

0.21

ä¸ĭçļĦ

0.19

 below

0.19

.Bottom

0.19

 Bottom

0.19

bottom

0.18

-down

0.18

-bottom

0.18

-level

0.18

below

0.17

Activations Density 0.196%

references to structural components and their positions within various contexts

No Comments

No Known Activations