INDEX

Explanations

concepts related to emotional pain and betrayal

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

Negative Logits

 Ð¾Ð¿Ð°Ñģ

-0.17

ihar

-0.16

 irritating

-0.16

 menace

-0.16

agen

-0.15

eken

-0.15

åį±

-0.15

_Static

-0.15

ermal

-0.15

 fatal

-0.15

POSITIVE LOGITS

 emotional

0.25

 trauma

0.23

 emotions

0.23

pts

0.23

Tra

0.22

tra

0.22

 pain

0.21

PT

0.21

 feelings

0.21

pts

0.21

Activations Density 0.493%

concepts related to emotional pain and betrayal

No Known Activations