INDEX

Explanations

discourse related to understanding and explanation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ainen

-0.17

ãģĵãģ¨ãģ§

-0.16

$_['

-0.16

 this

-0.15

 cela

-0.14

this

-0.14

illes

-0.14

 ours

-0.14

zcze

-0.14

Could

-0.13

POSITIVE LOGITS

 requires

0.39

requires

0.35

 Requires

0.30

 require

0.28

 must

0.25

Requires

0.25

must

0.24

 Require

0.24

 necesita

0.23

å¿ħé¡»

0.23

Activations Density 0.136%

discourse related to understanding and explanation

No Comments

No Known Activations