INDEX

Explanations

proper nouns and names related to notable individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

iks

-0.16

ymes

-0.16

è°·

-0.16

cak

-0.15

 destinationViewController

-0.15

éĢģæĸĻçĦ¡æĸĻ

-0.15

Ã¢b

-0.14

atori

-0.14

alion

-0.14

ousse

-0.14

POSITIVE LOGITS

 Habit

0.14

ped

0.13

raft

0.13

*s

0.13

FS

0.13

fs

0.13

 habit

0.13

antz

0.13

 himself

0.13

ain

0.13

Activations Density 0.078%

proper nouns and names related to notable individuals or entities

No Comments

No Known Activations