INDEX

Explanations

titles and positions of high-ranking officials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ocy

-0.15

lop

-0.14

ÎºÎµ

-0.14

duk

-0.14

lopen

-0.14

breadcrumb

-0.14

ncia

-0.14

ajs

-0.14

lied

-0.14

ndef

-0.14

POSITIVE LOGITS

pres

0.27

 president

0.24

-president

0.23

 present

0.21

-chair

0.21

 prÃ©s

0.20

 President

0.19

 pres

0.19

-pres

0.19

_pres

0.18

Activations Density 0.005%

titles and positions of high-ranking officials

No Comments

No Known Activations