INDEX

Explanations

programming-related syntax, specifically array or list indexing and assignments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

arto

-0.17

velle

-0.17

vre

-0.15

itzer

-0.15

lesc

-0.14

vou

-0.14

ETCH

-0.14

ÎºÎ¹

-0.14

yne

-0.14

 challenge

-0.14

POSITIVE LOGITS

eners

0.18

osate

0.15

ozor

0.15

ÌĢ

0.15

.GroupLayout

0.15

assage

0.14

lest

0.14

hek

0.14

stra

0.13

_asm

0.13

Activations Density 0.103%

programming-related syntax, specifically array or list indexing and assignments

No Comments

No Known Activations