INDEX

Explanations

titles of animated series or films

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

lom

-0.15

ãĥ¼ãĥŃ

-0.15

ambre

-0.15

bairro

-0.14

bable

-0.14

åŀĭ

-0.14

é¸¡

-0.14

oeff

-0.13

ampire

-0.13

POSITIVE LOGITS

arin

0.15

 Animator

0.15

owl

0.15

aln

0.14

oldown

0.14

 anim

0.14

olate

0.14

ows

0.14

 Disney

0.14

 vibr

0.14

Activations Density 0.143%

titles of animated series or films

No Comments

No Known Activations