INDEX

Explanations

interrogative phrases that solicit opinions or thoughts from the reader

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ceased

-0.15

teri

-0.15

etwork

-0.15

weetalert

-0.14

tik

-0.14

ä¹ĥ

-0.14

jez

-0.14

Ńå·ŀ

-0.14

Î¹

-0.14

 mappedBy

-0.13

POSITIVE LOGITS

 thinks

0.20

 think

0.20

onet

0.19

 opin

0.19

 mean

0.19

 Honest

0.18

emean

0.17

 pense

0.16

 opinions

0.16

atham

0.16

Activations Density 0.052%

interrogative phrases that solicit opinions or thoughts from the reader

No Comments

No Known Activations