INDEX

Explanations

occurrences of the word "destroyed."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

boru

-0.15

atte

-0.15

fax

-0.14

ãģĤãģ£ãģŁ

-0.14

utch

-0.14

ongoose

-0.13

boo

-0.13

Ð°Ð½Ð½Ñİ

-0.13

 Haram

-0.13

odes

-0.13

POSITIVE LOGITS

umer

0.19

laz

0.17

uide

0.15

avian

0.15

 stddev

0.15

pret

0.14

mall

0.14

Äĩi

0.14

.position

0.13

lst

0.13

Activations Density 0.009%

occurrences of the word "destroyed."

No Comments

No Known Activations