INDEX

Explanations

instances of negative feedback and mental health references

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

mga

-0.16

UNG

-0.15

äººä»¬

-0.15

IPs

-0.14

ungle

-0.14

 types

-0.14

sects

-0.14

 workforce

-0.14

 entries

-0.14

 various

-0.14

POSITIVE LOGITS

 item

0.28

 piece

0.26

 member

0.23

ä¹ĭä¸Ģ

0.22

 person

0.21

 instance

0.20

member

0.20

 molecule

0.20

 Piece

0.19

 element

0.18

Activations Density 1.195%

instances of negative feedback and mental health references

No Comments

No Known Activations