INDEX

Explanations

punctuation marks and isolated instances of certain non-zero words or phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

vens

-0.16

è°

-0.16

ÑĦÐ¾ÑĢÐ¼Ð¸

-0.15

anos

-0.14

woo

-0.14

ook

-0.14

nee

-0.14

ooks

-0.14

Ø¤ÙĪÙĦ

-0.13

aid

-0.13

POSITIVE LOGITS

omat

0.15

klad

0.14

 hepsi

0.14

INGTON

0.14

RIA

0.14

landa

0.13

zbollah

0.13

ellas

0.13

ths

0.13

ë¦¬ìķĦ

0.13

Activations Density 0.213%

punctuation marks and isolated instances of certain non-zero words or phrases

No Comments

No Known Activations