INDEX

Explanations

questions seeking clarification or information

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

oke

-0.17

ously

-0.15

 Holt

-0.14

 weren

-0.14

-switch

-0.13

 åĨĨ

-0.13

tran

-0.13

ucha

-0.12

Ð¾ÑĤÑĮ

-0.12

ADA

-0.12

POSITIVE LOGITS

soever

0.25

 exactly

0.24

 shall

0.24

 actually

0.23

 exact

0.19

 Shall

0.19

 Exactly

0.18

do

0.18

 should

0.18

are

0.18

Activations Density 0.127%

questions seeking clarification or information

No Comments

No Known Activations