INDEX

Explanations

concepts and terms related to monitoring and evaluation activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 strtok

-0.17

ÑĤÐ°Ð»

-0.15

haz

-0.15

vit

-0.14

 ìŀ¡

-0.14

 mold

-0.14

.invoke

-0.13

oenix

-0.13

sono

-0.13

 molded

-0.13

POSITIVE LOGITS

 evalu

0.27

Eval

0.27

Rig

0.26

 Evaluation

0.25

evaluation

0.24

 evaluation

0.24

 evaluator

0.24

Evaluation

0.24

 eval

0.23

 evaluations

0.23

Activations Density 0.027%

concepts and terms related to monitoring and evaluation activities

No Comments

No Known Activations

concepts and terms related to monitoring and evaluation activities

No Comments

No Known Activations