INDEX

Explanations

inquiries or questions regarding roles, statistics, or changes in a specific context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ovah

-0.17

allo

-0.15

avad

-0.14

ulant

-0.14

tÃŃ

-0.14

atters

-0.14

resse

-0.14

agation

-0.14

æĹ

-0.13

POSITIVE LOGITS

 role

0.19

 sort

0.18

 chance

0.17

 impact

0.16

 lesson

0.16

ä½ĵ

0.16

 sense

0.15

 effect

0.15

 precisely

0.15

 sorts

0.15

Activations Density 0.111%

inquiries or questions regarding roles, statistics, or changes in a specific context

No Comments

No Known Activations