INDEX

Explanations

phrases indicating quantity or comparative relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ouz

-0.16

 tolik

-0.15

hurst

-0.15

iza

-0.15

izu

-0.15

ruÅ¾

-0.14

ŀæĢ§

-0.14

isia

-0.14

ÑĨÑĥ

-0.14

astle

-0.14

POSITIVE LOGITS

as

0.29

 early

0.19

EAR

0.17

early

0.16

 ÐºÐ°Ðº

0.16

 dÃ¨s

0.15

encil

0.14

 Vine

0.14

æĹ©

0.14

EAR

0.14

Activations Density 0.021%

phrases indicating quantity or comparative relationships

No Comments

No Known Activations