INDEX

Explanations

issues related to validation errors and unexpected values in coded components

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

ittel

-0.15

åķı

-0.14

edula

-0.14

 Ð¿Ð¾Ð¶

-0.14

 scor

-0.14

éĥ

-0.13

 boil

-0.13

weet

-0.13

rect

-0.13

enko

-0.13

POSITIVE LOGITS

eren

0.16

0.15

èĬ¬

0.14

Ĥ¨

0.14

 Survival

0.14

Ð½ÑĸÑģÑĤ

0.14

Î¿Î¹

0.14

ãĤ¶ãĥ¼

0.14

isel

0.14

 multic

0.14

Activations Density 0.075%

issues related to validation errors and unexpected values in coded components

No Comments

No Known Activations