INDEX

Explanations

references to types of processes or treatments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

trie

-0.16

endra

-0.15

Unchecked

-0.15

berger

-0.14

anto

-0.14

ÑĢÐ°ÑģÑĤÐ°

-0.14

çĬ

-0.14

(stdin

-0.13

 Flour

-0.13

ande

-0.13

POSITIVE LOGITS

 various

0.33

 variety

0.30

 varied

0.29

 many

0.28

 Various

0.28

Various

0.27

 myriad

0.26

 numerous

0.25

ãģķãģ¾

0.25

 diverse

0.23

Activations Density 0.194%

references to types of processes or treatments

No Comments

No Known Activations