INDEX

Explanations

No Explanations Found

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

rak

-0.16

abei

-0.15

oreach

-0.15

aliases

-0.14

 learnt

-0.14

_blocking

-0.14

 bisher

-0.14

 kvinn

-0.13

POSITIVE LOGITS

acc

0.19

Acc

0.18

acc

0.17

ACC

0.16

_ACC

0.16

mob

0.16

Acc

0.15

 ordinary

0.15

ACC

0.15

 Ordinary

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.