INDEX

Explanations

log messages and error reports related to server operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 prÃ©c

-0.16

illis

-0.15

Ð¾Ð»Ð¾Ð¼

-0.14

gratis

-0.14

 Orient

-0.14

cre

-0.14

ipeg

-0.14

emente

-0.13

 foil

-0.13

omite

-0.13

POSITIVE LOGITS

aign

0.16

0.14

rist

0.14

coe

0.14

ORA

0.13

ÑĢÐ¾

0.13

_decor

0.13

cheon

0.13

ron

0.13

Activations Density 0.040%

log messages and error reports related to server operations

No Comments

No Known Activations