INDEX

Explanations

questions related to guidance and instructions

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

Ð¾ÑĩÐºÑĥ

-0.15

 Block

-0.15

awa

-0.14

èŃ·

-0.14

och

-0.14

ertz

-0.14

Haw

-0.14

 Castillo

-0.13

man

-0.13

verse

-0.13

POSITIVE LOGITS

iaux

0.20

eum

0.17

èĸ¦

0.16

ureau

0.15

eck

0.15

isphere

0.15

eci

0.14

isContained

0.14

urent

0.14

ì°½

0.14

Activations Density 0.010%

questions related to guidance and instructions

No Comments

No Known Activations