INDEX

Explanations

instances of contradiction or conflict in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 ONLINE

-0.15

vely

-0.15

alu

-0.14

enny

-0.14

 Online

-0.14

 Sink

-0.14

berapa

-0.14

à¤Ĥà¤ľà¤¨

-0.14

 haze

-0.13

ä»ĺ

-0.13

POSITIVE LOGITS

 also

0.23

too

0.20

 Ø£ÙĬØ¶

0.18

 indeed

0.17

 Ø£ÙĬØ¶Ø§

0.17

also

0.17

 ocksÃ¥

0.17

 tambiÃ©n

0.16

 herself

0.16

ä¹Ł

0.16

Activations Density 0.252%

instances of contradiction or conflict in narratives

No Comments

No Known Activations