INDEX

Explanations

expressions related to extreme negative outcomes or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

Î´Î¬

-0.19

-0.17

etty

-0.16

acock

-0.16

ØŃÙĩ

-0.16

'gc

-0.15

etur

-0.14

iele

-0.14

elsen

-0.14

agma

-0.14

POSITIVE LOGITS

 worse

0.20

 Worse

0.17

-case

0.15

 worst

0.15

 Wolfe

0.14

æ¼«

0.14

auga

0.14

ãĥ¼ãĤ¹ãĥĪ

0.13

JOB

0.13

ê

0.13

Activations Density 0.030%

expressions related to extreme negative outcomes or experiences

No Comments

No Known Activations