INDEX

Explanations

references to interpersonal interactions and achievements

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

_MACRO

-0.15

Via

-0.14

 Behavior

-0.14

 conduct

-0.14

rote

-0.14

icho

-0.13

oric

-0.13

elfast

-0.13

 behavior

-0.13

POSITIVE LOGITS

 faster

0.24

 beat

0.23

 ká»ĭp

0.22

 åħĪ

0.22

åħĪ

0.21

 intercept

0.20

 fastest

0.20

 trÆ°á»Ľc

0.20

 antes

0.20

 before

0.20

Activations Density 0.127%

references to interpersonal interactions and achievements

No Known Activations