INDEX

Explanations

negative connotations associated with incidents and accidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

astr

-0.18

URY

-0.15

 breat

-0.15

.vendor

-0.15

mj

-0.14

anzi

-0.14

cke

-0.13

è³Ģ

-0.13

Ø¹ÙĪØ¯

-0.13

vendor

-0.13

POSITIVE LOGITS

 Bloss

0.16

vale

0.15

 police

0.15

yled

0.15

 death

0.15

Į

0.14

á»įt

0.14

 Elaine

0.13

 events

0.13

 enough

0.13

Activations Density 1.007%

negative connotations associated with incidents and accidents

No Comments

No Known Activations