INDEX

Explanations

food-related terms and cooking instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

isel

-0.17

agli

-0.15

quent

-0.14

achel

-0.14

hot

-0.14

//{{

-0.14

ricanes

-0.13

ollen

-0.13

Unsigned

-0.13

EventListener

-0.13

POSITIVE LOGITS

 Weinstein

0.19

ERO

0.16

HEET

0.15

rosse

0.14

yan

0.14

Ø§ÛĮØ±

0.14

hee

0.14

enne

0.13

.foundation

0.13

pob

0.13

Activations Density 0.039%

food-related terms and cooking instructions

No Comments

No Known Activations