INDEX

Explanations

references to tables and structured information

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 aren

-0.18

morph

-0.18

 morph

-0.18

ayah

-0.17

gaard

-0.16

æĸ

-0.15

 Aren

-0.15

 weren

-0.14

 hadn

-0.14

 Morph

-0.14

POSITIVE LOGITS

 bids

0.22

sav

0.22

 condu

0.18

bad

0.17

 milit

0.17

bid

0.17

 teen

0.17

 afford

0.17

 answers

0.17

 furn

0.17

Activations Density 0.222%

references to tables and structured information

No Known Activations