INDEX

Explanations

terms related to confirmation of existing knowledge or facts

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

_AA

-0.16

 cele

-0.15

Ø§Ø¨Ø±

-0.15

ereum

-0.14

 Cele

-0.14

anguage

-0.14

 celebrity

-0.14

alien

-0.14

alone

-0.14

anggal

-0.13

POSITIVE LOGITS

 unknown

0.33

unknown

0.26

 Unknown

0.26

_unknown

0.24

Unknown

0.23

 UNKNOWN

0.23

UNKNOWN

0.20

 initial

0.18

unks

0.18

 undefined

0.18

Activations Density 0.005%

terms related to confirmation of existing knowledge or facts

No Comments

No Known Activations