INDEX

Explanations

references to IT-related terms and technology concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

hound

-0.21

hire

-0.20

hill

-0.20

houses

-0.20

hic

-0.19

halt

-0.18

hid

-0.18

eson

-0.18

hair

-0.18

hana

-0.18

POSITIVE LOGITS

inerary

0.26

zen

0.23

ting

0.22

tees

0.22

TY

0.21

TING

0.21

teenth

0.20

hec

0.20

tee

0.20

izens

0.20

Activations Density 0.111%

references to IT-related terms and technology concepts

No Comments

No Known Activations