INDEX

Explanations

mathematical notation and expressions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

umerator

-0.17

{{{

-0.16

ofile

-0.16

erland

-0.16

fault

-0.15

chrift

-0.14

ochen

-0.14

Ø§ÛĮØ±

-0.14

ients

-0.14

leton

-0.14

POSITIVE LOGITS

zelf

0.18

HING

0.14

 fair

0.14

 resume

0.13

 nerv

0.13

EB

0.13

agogue

0.13

_registro

0.13

ly

0.13

aker

0.13

Activations Density 0.075%

mathematical notation and expressions

No Comments

No Known Activations