INDEX

Explanations

structured sequences and patterns within numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ordon

-0.16

 brushes

-0.15

ForRow

-0.15

oad

-0.14

 vict

-0.14

Ø¨Ø§ÙĨ

-0.14

pa

-0.14

sw

-0.14

Ã¼ss

-0.14

Re

-0.13

POSITIVE LOGITS

ï»¿#

0.18

kowski

0.17

¤í

0.15

pest

0.15

 å¥

0.14

enaire

0.14

'''č↵

0.14

 testim

0.14

Ð»ÐµÐ¼

0.13

NECT

0.13

Activations Density 0.110%

structured sequences and patterns within numerical data

No Comments

No Known Activations