INDEX

Explanations

negative values or indicators of failure or loss

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ï»¿using

-0.17

----------</

-0.17

-------------</

-0.16

aversal

-0.16

-|

-0.15

EMPL

-0.15

-',

-0.15

-et

-0.15

foy

-0.15

-</

-0.15

POSITIVE LOGITS

ve

0.31

/+

0.28

webkit

0.27

ve

0.24

Infinity

0.20

Ve

0.20

moz

0.20

_ve

0.19

0.18

Activations Density 0.063%

negative values or indicators of failure or loss

No Comments

No Known Activations