INDEX

Explanations

phrases that contrast knowledge with experience

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

Negative Logits

ehler

-0.15

à¹īà¸Ħ

-0.14

ãĤ¤ãĥĦ

-0.14

chner

-0.14

igg

-0.14

bourne

-0.14

asley

-0.14

ãĥ¼ãĥĢ

-0.13

ying

-0.13

ÐºÐ¾ÑĤ

-0.13

POSITIVE LOGITS

tun

0.15

tatus

0.14

 Rogers

0.14

Kad

0.14

tuk

0.14

lad

0.14

 Moses

0.14

 Owens

0.14

Smy

0.14

Activations Density 0.298%

phrases that contrast knowledge with experience

No Known Activations