INDEX

Explanations

terms and metrics related to model evaluation and optimization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 cred

-0.16

mat

-0.15

 Ù¾ÙĨ

-0.15

 neutral

-0.14

ãģļ

-0.14

em

-0.14

ardon

-0.14

 disc

-0.14

 åıĳ

-0.14

æľīçļĦ

-0.13

POSITIVE LOGITS

 Welch

0.19

onu

0.17

asca

0.16

SSERT

0.16

elters

0.15

ç«ł

0.14

ertz

0.14

StackNavigator

0.14

loser

0.14

obraz

0.14

Activations Density 0.079%

terms and metrics related to model evaluation and optimization

No Comments

No Known Activations