INDEX

Explanations

punctuation marks and spacing elements in the text

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

abs

-0.15

ãĥ³ãĥĳ

-0.15

 disag

-0.14

ould

-0.14

izard

-0.14

efeller

-0.14

BindingUtil

-0.14

izza

-0.14

.directive

-0.14

Ë

-0.14

POSITIVE LOGITS

adro

0.16

 Kraft

0.15

iah

0.14

AVED

0.14

ed

0.14

alg

0.13

 Thought

0.13

ismet

0.13

asha

0.13

ohn

0.13

Activations Density 0.045%

punctuation marks and spacing elements in the text

No Comments

No Known Activations