INDEX

Explanations

key terms related to political or governmental activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

hte

-0.16

inch

-0.16

ateur

-0.16

Ñī

-0.16

branch

-0.15

ayet

-0.14

ISCO

-0.14

zych

-0.14

empor

-0.14

POSITIVE LOGITS

 Quad

0.21

 India

0.20

 Indian

0.20

 External

0.20

New

0.19

 Indians

0.18

Quad

0.18

Dok

0.18

Lad

0.17

MEA

0.17

Activations Density 0.015%

key terms related to political or governmental activities

No Comments

No Known Activations

key terms related to political or governmental activities

No Comments

No Known Activations