INDEX

Explanations

references to Canada and Canadian identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

yun

-0.16

fully

-0.15

sm

-0.15

kh

-0.15

omor

-0.14

ech

-0.14

kr

-0.14

ãģ¨ãģĵãĤį

-0.14

sel

-0.14

POSITIVE LOGITS

igua

0.26

 Goose

0.24

adians

0.24

.gc

0.21

adian

0.20

 goose

0.19

 provinces

0.18

á»¥

0.18

berra

0.18

 Trudeau

0.18

Activations Density 0.039%

references to Canada and Canadian identity

No Comments

No Known Activations