INDEX

Explanations

references to previously established or current entities and systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ings

-0.21

 newly

-0.17

 modern

-0.15

new

-0.15

 recent

-0.14

 latest

-0.14

headed

-0.14

 former

-0.14

owl

-0.14

POSITIVE LOGITS

/current

0.25

/original

0.23

/new

0.20

 ones

0.19

 Ones

0.17

-established

0.16

-generation

0.15

/up

0.15

tlement

0.15

zer

0.15

Activations Density 0.030%

references to previously established or current entities and systems

No Comments

No Known Activations