INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

lb

-0.16

ÚĨÛĮ

-0.15

ernet

-0.15

iez

-0.15

Å¾il

-0.14

ermann

-0.14

oller

-0.14

 PyTuple

-0.14

ÑĢÐµÐ¶

-0.14

achel

-0.14

POSITIVE LOGITS

entrant

0.14

eson

0.13

 Chick

0.13

ycin

0.13

leveland

0.13

_keeper

0.13

 Helm

0.13

ÐºÐ¸

0.13

Tes

0.12

 Terraria

0.12

Activations Density 0.000%

No Known Activations

This feature has no known activations.