INDEX

Explanations

messages and statements that convey important ideas or concerns

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

untas

-0.17

pector

-0.14

uco

-0.14

à¤Ĥà¤§

-0.14

HING

-0.14

_ptrs

-0.14

ics

-0.14

åĿ

-0.14

spec

-0.13

Pointer

-0.13

POSITIVE LOGITS

 message

0.35

message

0.29

 Message

0.28

 messages

0.27

 convey

0.25

-message

0.25

/message

0.25

 loud

0.24

 MESSAGE

0.24

(message

0.24

Activations Density 0.065%

messages and statements that convey important ideas or concerns

No Comments

No Known Activations