INDEX

Explanations

references to iconic or well-known cultural symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

chop

-0.18

-0.16

 newest

-0.15

Ð¸ÑĤÐ¸

-0.14

 Wonderland

-0.13

rica

-0.13

aturas

-0.13

orda

-0.13

 latest

-0.13

á»ķ

-0.13

POSITIVE LOGITS

ingles

0.18

/pop

0.15

imum

0.15

/original

0.15

kov

0.15

ummer

0.14

wy

0.14

/icon

0.14

overn

0.14

ä¸Ģåº¦

0.14

Activations Density 0.048%

references to iconic or well-known cultural symbols

No Comments

No Known Activations