INDEX

Explanations

references to specific movies and their elements

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

peater

-0.16

ollah

-0.16

kus

-0.15

å¤§åħ¨

-0.15

uchs

-0.14

å°Ħ

-0.14

lÃ¡

-0.14

pletion

-0.13

 meaningless

-0.13

åįĴ

-0.13

POSITIVE LOGITS

 historical

0.37

 real

0.34

 Historical

0.33

 factual

0.30

bi

0.28

 true

0.27

histor

0.27

 fictional

0.27

true

0.26

real

0.26

Activations Density 0.107%

references to specific movies and their elements

No Comments

No Known Activations