INDEX

Explanations

patterns related to programming functions and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

lah

-0.17

 Choices

-0.16

ysl

-0.16

aphrag

-0.15

ALAR

-0.15

isoft

-0.15

cai

-0.15

 jenter

-0.15

Choices

-0.15

caff

-0.14

POSITIVE LOGITS

 check

0.22

 checks

0.19

 Judge

0.19

 checking

0.19

 test

0.18

åĪ¤æĸŃ

0.18

 StringComparison

0.18

check

0.18

 judge

0.17

 Check

0.17

Activations Density 0.483%

patterns related to programming functions and structures

No Comments

No Known Activations