INDEX

Explanations

individuals' names and their corresponding statements or opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ã¡p

-0.15

adal

-0.15

 Serge

-0.14

entes

-0.14

imas

-0.14

oog

-0.14

onium

-0.14

VRT

-0.14

 Strauss

-0.14

AMED

-0.14

POSITIVE LOGITS

 aborted

0.15

inan

0.14

0.13

iddi

0.13

-pass

0.13

lesi

0.13

Scenario

0.13

ç¿

0.13

ipay

0.13

Activations Density 0.019%

individuals' names and their corresponding statements or opinions

No Comments

No Known Activations