INDEX

Explanations

references to people and their roles or contributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÑĩÐµÑģÐºÐ¾Ðµ

-0.16

aja

-0.16

atto

-0.15

ambique

-0.15

heed

-0.14

á»ĩn

-0.14

aData

-0.14

Éµ

-0.14

arc

-0.14

rado

-0.14

POSITIVE LOGITS

ilarity

0.14

Ø³Ø§Ø¨

0.14

ont

0.14

ulen

0.14

esser

0.14

orial

0.13

 Forgot

0.13

unt

0.13

ä½ľä¸º

0.13

æĵį

0.13

Activations Density 0.096%

references to people and their roles or contributions

No Comments

No Known Activations