INDEX

Explanations

phrases related to assessments and evaluations of performance or quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

ombat

-0.16

ROID

-0.15

 à¤ªà¤¡

-0.14

icÃŃ

-0.14

ibus

-0.14

âĶ¬

-0.14

Bod

-0.14

roid

-0.13

iphone

-0.13

odi

-0.13

POSITIVE LOGITS

DISABLE

0.15

 Herrera

0.14

 unprotected

0.14

uw

0.14

alted

0.14

.Cmd

0.14

 consid

0.14

ModelProperty

0.14

spring

0.14

icts

0.13

Activations Density 0.111%

phrases related to assessments and evaluations of performance or quality

No Comments

No Known Activations