INDEX

Explanations

specific names and references related to authorship and publication details

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

$MESS

-0.16

rav

-0.14

antt

-0.14

stem

-0.14

aceut

-0.14

iek

-0.14

kli

-0.14

adiens

-0.14

 Bray

-0.14

loquent

-0.14

POSITIVE LOGITS

eks

0.16

oj

0.15

asley

0.14

erp

0.13

ewire

0.13

ered

0.13

(SIG

0.13

 Salon

0.13

 enrol

0.13

erialized

0.12

Activations Density 0.112%

specific names and references related to authorship and publication details

No Comments

No Known Activations