INDEX

Explanations

references to medium-sized entities or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ment

-0.18

ides

-0.16

zym

-0.16

iaux

-0.15

anter

-0.15

adi

-0.15

anal

-0.15

actory

-0.14

isd

-0.14

radi

-0.14

POSITIVE LOGITS

 sized

0.41

-sized

0.41

 Sized

0.31

-size

0.28

size

0.24

ship

0.23

-large

0.23

 size

0.22

-length

0.22

-range

0.22

Activations Density 0.019%

references to medium-sized entities or classifications

No Comments

No Known Activations