INDEX

Explanations

programming-related terms and functions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

cki

-0.17

.RunWith

-0.15

 stoi

-0.14

 migrationBuilder

-0.14

azu

-0.14

orda

-0.14

URED

-0.14

//=

-0.13

classnames

-0.13

PEED

-0.13

POSITIVE LOGITS

uja

0.14

 Bird

0.14

 Calif

0.14

_bed

0.14

Bird

0.14

erval

0.14

bron

0.14

ç·

0.13

ourd

0.13

 soud

0.13

Activations Density 0.081%

programming-related terms and functions

No Comments

No Known Activations