INDEX

Explanations

forms of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ëŀĳ

-0.14

 Heal

-0.14

ÑĥÐ¼Ð°

-0.14

pectral

-0.14

quat

-0.14

aise

-0.13

á»§ng

-0.13

tabs

-0.13

ynch

-0.13

Sm

-0.13

POSITIVE LOGITS

ìĿ´ìĸ´

0.17

/goto

0.15

itals

0.15

imar

0.14

onia

0.14

loser

0.14

addir

0.14

ç»Ĩ

0.13

Pai

0.13

geber

0.13

Activations Density 0.054%

forms of the verb "to be."

No Comments

No Known Activations