INDEX

Explanations

references to underperforming teams or negative statistics in sports contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

agos

-0.15

ØªÙĪÙĨ

-0.15

viso

-0.15

 jenter

-0.14

 Conc

-0.14

oub

-0.14

 Administr

-0.14

Ä±lÄ±Ã§

-0.14

icht

-0.14

æŃ

-0.14

POSITIVE LOGITS

 Feinstein

0.16

oral

0.15

isser

0.15

orer

0.15

uman

0.15

 Garten

0.14

 Angeles

0.14

Ð¾ÐºÐµ

0.14

etable

0.14

ÑģÐ»Ð¸

0.13

Activations Density 0.747%

references to underperforming teams or negative statistics in sports contexts

No Comments

No Known Activations