INDEX

Explanations

keywords related to presentations and evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 yerde

-0.15

pez

-0.15

izontal

-0.14

isoner

-0.14

laz

-0.14

ĭè¯ķ

-0.14

atra

-0.14

 unnatural

-0.14

áº¡

-0.14

bable

-0.14

POSITIVE LOGITS

Rel

0.16

hi

0.16

leigh

0.15

usters

0.14

eca

0.14

 nursing

0.14

rust

0.14

ÅĤo

0.14

Rin

0.14

Activations Density 0.007%

keywords related to presentations and evaluations

No Comments

No Known Activations