INDEX

Explanations

terms related to data evaluation and metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 fasc

-0.16

_capabilities

-0.15

 drag

-0.14

ãĥĲãĥ¼

-0.14

fal

-0.14

ä¸²

-0.14

anova

-0.14

iston

-0.13

å«Į

-0.13

enschaft

-0.13

POSITIVE LOGITS

 dataset

0.39

 Dataset

0.33

 datasets

0.32

dataset

0.31

Dataset

0.28

 training

0.28

 benchmark

0.28

benchmark

0.28

_dataset

0.26

 train

0.25

Activations Density 0.075%

terms related to data evaluation and metrics

No Comments

No Known Activations