INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

implify

-0.18

sm

-0.16

atab

-0.15

osphere

-0.15

aze

-0.15

abbo

-0.15

ovalo

-0.15

kat

-0.15

lette

-0.14

POSITIVE LOGITS

iani

0.18

estre

0.16

hlen

0.15

_HEAP

0.15

reed

0.15

culus

0.14

 Pret

0.14

okus

0.13

licit

0.13

lsen

0.13

Activations Density 0.002%

No Comments

No Known Activations