INDEX

Explanations

references to the audience or audience members in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ï¸ı

-0.20

thal

-0.18

ides

-0.16

ocal

-0.15

gor

-0.15

nown

-0.15

ãģ¾ãģŁ

-0.15

ica

-0.15

ivot

-0.15

deg

-0.14

POSITIVE LOGITS

 members

0.30

members

0.28

 member

0.27

member

0.26

-member

0.23

hip

0.21

_member

0.21

/view

0.20

/client

0.20

/customer

0.20

Activations Density 0.016%

references to the audience or audience members in various contexts

No Comments

No Known Activations