INDEX

Explanations

architectural features and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Ð»ÐµÐ¹

-0.15

AKER

-0.15

Specifier

-0.15

chner

-0.15

InvalidOperationException

-0.14

imbus

-0.14

åĳĬ

-0.14

 poil

-0.14

itemap

-0.14

airo

-0.14

POSITIVE LOGITS

 intact

0.15

anne

0.14

Vi

0.14

alse

0.14

ages

0.14

 fact

0.14

 reput

0.14

xt

0.14

ullen

0.14

ona

0.14

Activations Density 0.033%

architectural features and structures

No Comments

No Known Activations