INDEX

Explanations

characters or symbols that appear frequently

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

toy

-0.17

mo

-0.16

enan

-0.15

(er

-0.15

er

-0.15

uida

-0.15

 å¸Ĥ

-0.14

Nov

-0.14

toy

-0.14

POSITIVE LOGITS

let

0.20

js

0.20

lett

0.20

bred

0.20

leted

0.20

lete

0.20

lesen

0.20

rint

0.19

sz

0.19

rette

0.19

Activations Density 0.003%

characters or symbols that appear frequently

No Comments

No Known Activations