INDEX

Explanations

references to reactions and interactions among people and organizations regarding controversial topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 mostly

-0.23

 vÄĽtÅ¡inou

-0.23

 generally

-0.22

 obvykle

-0.21

 usually

-0.21

usually

-0.21

 souvent

-0.21

 often

-0.21

mostly

-0.20

 Ð¾Ð±ÑĭÑĩÐ½Ð¾

-0.20

POSITIVE LOGITS

 even

0.28

may

0.28

even

0.24

 simply

0.24

 sogar

0.23

 outright

0.22

 might

0.22

çĶļèĩ³

0.21

may

0.21

 downright

0.21

Activations Density 0.207%

references to reactions and interactions among people and organizations regarding controversial topics

No Comments

No Known Activations