INDEX

Explanations

Boolean values and attributes in code

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

Ð°ÑĢÐ°ÑĤ

-0.16

oll

-0.16

naire

-0.15

lot

-0.15

esc

-0.15

reas

-0.15

richt

-0.14

etag

-0.14

ecs

-0.14

.resume

-0.14

POSITIVE LOGITS

/false

0.23

ushima

0.19

oplast

0.16

izoph

0.15

0.14

opher

0.14

orton

0.14

Tru

0.14

orida

0.14

ellido

0.14

Activations Density 0.047%

Boolean values and attributes in code

No Comments

No Known Activations