INDEX

Explanations

references to first events, innovations, or products in various contexts

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

á»ĳt

-0.18

åıĪ

-0.17

æīįèĥ½

-0.17

 weiter

-0.17

ulumi

-0.16

 Various

-0.16

ibel

-0.15

akan

-0.15

æīį

-0.14

 various

-0.14

POSITIVE LOGITS

 ever

0.59

 EVER

0.46

-ever

0.44

ever

0.42

 anywhere

0.37

Ever

0.32

 Ever

0.31

 since

0.29

EVER

0.29

 jamais

0.28

Activations Density 0.267%

references to first events, innovations, or products in various contexts

No Comments

No Known Activations