INDEX

Explanations

references to past events and actions related to discussions and conclusions

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

elop

-0.15

 wives

-0.13

erna

-0.13

wives

-0.13

cial

-0.13

ØªØ§Ø¨

-0.13

 daughters

-0.13

.Adam

-0.12

luÅ¾

-0.12

ucked

-0.12

POSITIVE LOGITS

ol

0.32

ole

0.25

Mr

0.24

 poor

0.23

 dear

0.23

old

0.22

mr

0.21

OUR

0.21

 Herr

0.20

Mr

0.20

Activations Density 0.267%

references to past events and actions related to discussions and conclusions

No Comments

No Known Activations