INDEX

Explanations

pieces of code defining interfaces and classes in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

.her

-0.16

ugi

-0.15

Toy

-0.15

(UINT

-0.14

dam

-0.14

omba

-0.14

Vy

-0.14

PLY

-0.13

her

-0.13

 Vest

-0.13

POSITIVE LOGITS

slash

0.17

ì»

0.16

avaÅŁ

0.15

Det

0.15

lash

0.15

urgeon

0.14

 PROGMEM

0.14

aload

0.14

WC

0.14

arat

0.14

Activations Density 0.065%

pieces of code defining interfaces and classes in a programming context

No Comments

No Known Activations