INDEX

Explanations

product-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

iah

-0.16

emas

-0.15

auge

-0.15

 handheld

-0.14

ema

-0.14

cil

-0.14

_probe

-0.14

Rew

-0.14

 Heath

-0.13

 portable

-0.13

POSITIVE LOGITS

 case

0.29

 Cases

0.27

 cases

0.25

 Case

0.24

Case

0.23

ãĤ±ãĥ¼ãĤ¹

0.23

Cases

0.22

case

0.22

 caso

0.21

 casos

0.20

Activations Density 0.015%

product-related terminology

No Comments

No Known Activations