INDEX

Explanations

a wide variety of common conjunctions and modifiers that connect ideas in a text

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

ä»²

-0.18

ÙĨØ§ÙĨ

-0.16

enou

-0.15

itele

-0.15

riz

-0.14

ellular

-0.14

auer

-0.14

 Academy

-0.14

adium

-0.14

nero

-0.14

POSITIVE LOGITS

ternet

0.15

/cache

0.14

acie

0.14

IGIN

0.14

ç¯Ģ

0.13

 jadx

0.13

irit

0.13

ãĤ¹ãĥŀ

0.13

erule

0.13

 poÅĻad

0.13

Activations Density 0.005%

a wide variety of common conjunctions and modifiers that connect ideas in a text

No Comments

No Known Activations