INDEX

Explanations

references to rankings or positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

andle

-0.16

anus

-0.14

YN

-0.14

iaux

-0.14

olare

-0.14

ijke

-0.13

tph

-0.13

 èĲ

-0.13

priority

-0.13

ynch

-0.13

POSITIVE LOGITS

 ones

0.19

 stage

0.17

ly

0.17

rtl

0.16

 Ones

0.15

.ci

0.15

agon

0.15

 alike

0.15

ledge

0.15

.RightToLeft

0.15

Activations Density 0.087%

references to rankings or positions

No Comments

No Known Activations