INDEX

Explanations

questions starting with "why."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ffa

-0.16

onse

-0.15

amework

-0.15

leine

-0.14

sters

-0.14

UNET

-0.14

ners

-0.14

ivate

-0.14

aurus

-0.14

wins

-0.13

POSITIVE LOGITS

ever

0.23

alla

0.19

do

0.19

did

0.18

 does

0.17

te

0.17

 bother

0.16

 else

0.16

 waste

0.16

not

0.16

Activations Density 0.023%

questions starting with "why."

No Comments

No Known Activations