INDEX

Explanations

dialogue and conversational responses

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

 nepÅĻ

-0.14

 gros

-0.14

/command

-0.14

ody

-0.14

base

-0.14

amburger

-0.14

åķı

-0.14

 Base

-0.14

/tag

-0.14

ableObject

-0.14

POSITIVE LOGITS

 added

0.19

 further

0.18

endregion

0.17

 endregion

0.16

ToAdd

0.16

angkan

0.16

-added

0.16

 adding

0.15

 addItem

0.15

added

0.15

Activations Density 0.057%

dialogue and conversational responses

No Comments

No Known Activations