INDEX

Explanations

intelligent and wise descriptors related to people or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

amespace

-0.17

Ìĥ

-0.16

bsite

-0.16

AZY

-0.16

ãģ£ãģį

-0.16

istrovstvÃŃ

-0.16

ainen

-0.16

ataire

-0.15

aday

-0.14

/by

-0.14

POSITIVE LOGITS

 minds

0.23

ale

0.23

 decisions

0.20

yp

0.20

 decision

0.19

acre

0.18

 enough

0.18

Ale

0.17

 thinking

0.17

ale

0.17

Activations Density 0.084%

intelligent and wise descriptors related to people or concepts

No Comments

No Known Activations