INDEX

Explanations

references to geographical locations or positional terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

Hob

-0.17

aura

-0.14

 structure

-0.14

ering

-0.14

 forward

-0.14

dom

-0.14

 function

-0.14

 Ð½Ð¸

-0.14

 mechanism

-0.14

/fixtures

-0.14

POSITIVE LOGITS

langs

0.20

wu

0.18

celik

0.18

fore

0.18

bei

0.15

Ã¼ber

0.15

esser

0.15

oder

0.15

Stick

0.15

ufen

0.15

Activations Density 0.022%

references to geographical locations or positional terms

No Comments

No Known Activations