INDEX

Explanations

phrases related to various forms of evaluation or judgment

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

 slew

-0.17

 Emmy

-0.16

 legit

-0.15

 convo

-0.15

chu

-0.15

 herk

-0.14

 Spike

-0.14

akis

-0.14

ces

-0.14

POSITIVE LOGITS

uster

0.18

ijd

0.17

alah

0.17

à¸±à¸Ļà¸Ļ

0.16

æľŃ

0.14

jot

0.14

 abort

0.14

oulder

0.14

iband

0.14

å¤§ä¼ļ

0.14

Activations Density 0.342%

phrases related to various forms of evaluation or judgment

No Comments

No Known Activations