INDEX

Explanations

programming-related syntax and commands

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

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

Raw

-0.18

{:

-0.17

RAW

-0.16

 tails

-0.15

essional

-0.15

esk

-0.15

ascal

-0.15

raw

-0.14

/vnd

-0.14

 Tail

-0.14

POSITIVE LOGITS

Wall

0.24

Winvalid

0.22

ped

0.20

pipe

0.20

ped

0.19

std

0.19

 Wall

0.19

stdlib

0.18

mav

0.18

DN

0.17

Activations Density 0.005%

programming-related syntax and commands

No Comments

No Known Activations