INDEX

Explanations

references to political questioning and accountability

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

 Norris

-0.17

oller

-0.16

iler

-0.15

Islam

-0.15

Ð°ÑĢÐ°ÑĤ

-0.14

ez

-0.14

ponsible

-0.14

 mosque

-0.14

ÏģÎ¯ÏĤ

-0.14

 ë¶Ļ

-0.13

POSITIVE LOGITS

 Hunter

0.38

Hunter

0.33

 hunter

0.30

 Rudy

0.27

Bur

0.27

 Ukrainian

0.25

 Ukraine

0.25

 Giuliani

0.24

 Hunters

0.24

Joe

0.23

Activations Density 0.003%

references to political questioning and accountability

No Comments

No Known Activations

references to political questioning and accountability

No Comments

No Known Activations