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

 simply

-0.19

 darn

-0.18

Simply

-0.16

inde

-0.16

 Simply

-0.16

YLON

-0.15

 Heck

-0.15

 indeed

-0.15

vider

-0.15

odies

-0.14

POSITIVE LOGITS

 tour

0.17

 Practices

0.15

 Practice

0.15

å½

0.15

 Literal

0.15

tour

0.14

 fucked

0.14

 practicing

0.14

 practice

0.14

 Prescott

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.