INDEX

Explanations

references to specific individuals or entities in a historical or athletic context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

acid

-0.16

agues

-0.15

getic

-0.15

antes

-0.14

Ĥ

-0.14

ë¯

-0.14

aters

-0.14

agn

-0.14

 Stam

-0.13

POSITIVE LOGITS

Ä±lÄ±Ã§

0.19

Ã¨ge

0.17

ÐµÐ½Ð·

0.17

ARGER

0.16

xes

0.15

encil

0.15

 devoted

0.15

UGE

0.15

tw

0.14

elihood

0.14

Activations Density 0.036%

references to specific individuals or entities in a historical or athletic context

No Comments

No Known Activations