INDEX

Explanations

instances of personal reflection and emotional responses in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

fir

-0.16

nock

-0.16

olute

-0.15

filer

-0.15

@mail

-0.15

uri

-0.15

_asm

-0.15

cha

-0.15

usan

-0.15

lez

-0.14

POSITIVE LOGITS

 hadn

0.36

 expected

0.36

 wasn

0.29

expected

0.29

 Expected

0.28

Expected

0.27

 never

0.27

 originally

0.27

 thought

0.25

 anticipated

0.25

Activations Density 0.282%

instances of personal reflection and emotional responses in the text

No Comments

No Known Activations