INDEX

Explanations

references to actions or events related to movement or change

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

ds

-0.16

 ourselves

-0.15

 ÙĪÙĦØ§

-0.15

 stuff

-0.15

immel

-0.14

éĤ£äºĽ

-0.14

os

-0.14

Its

-0.14

Ø±Ø¨Ùĩ

-0.13

Its

-0.13

POSITIVE LOGITS

ï»¿#

0.19

 proceedings

0.18

 emploi

0.16

edList

0.15

CCI

0.15

 whats

0.14

enko

0.14

ÐĲÑĢÑħÑĸÐ²

0.14

ï»¿↵↵

0.14

olest

0.14

Activations Density 0.340%

references to actions or events related to movement or change

No Comments

No Known Activations