INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

.Modules

-0.15

Managed

-0.14

 Managed

-0.14

upe

-0.14

qli

-0.13

aign

-0.13

 Jonas

-0.13

jong

-0.13

dna

-0.13

zier

-0.13

POSITIVE LOGITS

obs

0.25

observations

0.22

od

0.22

Observ

0.22

 observations

0.21

.obs

0.21

 Observation

0.21

_mr

0.21

 Observ

0.21

Obs

0.20

Activations Density 0.000%

No Known Activations

This feature has no known activations.