INDEX

Explanations

specific names and references linked to music, people, and notable events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Î½Î·

-0.15

 nervous

-0.15

noun

-0.15

 jedn

-0.15

Looper

-0.15

Ð´Ñĥ

-0.14

olik

-0.14

 ÄĮech

-0.14

 Karn

-0.14

erde

-0.14

POSITIVE LOGITS

(N

0.23

(NS

0.21

=N

0.16

@n

0.16

(Network

0.16

[N

0.15

,N

0.15

/N

0.15

>N

0.15

theless

0.15

Activations Density 0.121%

specific names and references linked to music, people, and notable events

No Comments

No Known Activations