INDEX

Explanations

references to classes and structures in a coding context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

-0.16

adal

-0.16

trand

-0.15

oldem

-0.15

acades

-0.14

etur

-0.14

amburger

-0.14

ableViewController

-0.14

iones

-0.13

POSITIVE LOGITS

apia

0.17

izer

0.15

rek

0.15

NC

0.14

unks

0.14

anko

0.14

ssi

0.14

 herk

0.14

pillar

0.14

lia

0.14

Activations Density 0.012%

references to classes and structures in a coding context

No Comments

No Known Activations