INDEX

Explanations

instances of the word "listen" and its variations, indicating a focus on auditory attention or engagement

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

ej

-0.16

ongan

-0.15

apo

-0.15

ulent

-0.15

ulu

-0.15

cam

-0.15

quires

-0.15

ãĥ³ãĤ¬

-0.15

acers

-0.14

ÑĢÐ¾Ñĩ

-0.14

POSITIVE LOGITS

/watch

0.25

AndServe

0.24

ening

0.22

 closely

0.20

/view

0.19

/read

0.18

eners

0.18

 ears

0.18

 carefully

0.18

 comprehension

0.17

Activations Density 0.022%

instances of the word "listen" and its variations, indicating a focus on auditory attention or engagement

No Comments

No Known Activations