INDEX

Explanations

mentions of political discussions and agreements

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

try

-0.16

çĪ±

-0.15

Abr

-0.15

 advised

-0.14

 even

-0.14

 excit

-0.14

 cooler

-0.14

Ã³st

-0.14

ardon

-0.14

argon

-0.13

POSITIVE LOGITS

dep

0.23

 welcome

0.23

 welcomes

0.22

welcome

0.21

 expresses

0.20

 welcomed

0.20

 express

0.19

 underline

0.19

express

0.18

 Welcome

0.18

Activations Density 0.055%

mentions of political discussions and agreements

No Comments

No Known Activations