INDEX

Explanations

phrases indicating causation or reason, particularly those starting with "due."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

-0.16

mony

-0.15

Ã´me

-0.15

 åĺ

-0.15

sh

-0.15

Î¯Î±

-0.15

iah

-0.14

rouw

-0.14

ÑĢÑĥÑĩ

-0.14

Ð¾ÑĢÐ¾Ð¶

-0.14

POSITIVE LOGITS

Dil

0.35

 diligence

0.28

dil

0.24

 dilig

0.21

Ã±as

0.21

ogh

0.19

 process

0.18

 dates

0.18

 date

0.18

-process

0.17

Activations Density 0.021%

phrases indicating causation or reason, particularly those starting with "due."

No Comments

No Known Activations