INDEX

Explanations

numerical data, particularly dates, years, and counts related to events or records

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ÑģÐºÐ¾ÑĢ

-0.15

gÄ±n

-0.14

ucher

-0.14

 infr

-0.14

.eof

-0.13

stdafx

-0.13

rances

-0.13

çĻ

-0.13

eam

-0.13

Alv

-0.13

POSITIVE LOGITS

eling

0.17

 Shah

0.16

azine

0.16

eks

0.15

obe

0.15

Parms

0.15

outu

0.14

ylland

0.14

.openg

0.14

TERN

0.13

Activations Density 0.044%

numerical data, particularly dates, years, and counts related to events or records

No Comments

No Known Activations