INDEX

Explanations

references to numbers and quantities in the text

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

isque

-0.17

EDIA

-0.16

.viewer

-0.15

edia

-0.15

 Bris

-0.14

Nim

-0.14

idos

-0.14

oker

-0.14

rosse

-0.14

anko

-0.14

POSITIVE LOGITS

kili

0.17

å¥ĩ

0.16

annels

0.15

upert

0.15

anel

0.15

ateria

0.14

ÙĥØª

0.14

 Clair

0.13

Activations Density 0.124%

references to numbers and quantities in the text

No Comments

No Known Activations