INDEX

Explanations

references to programming modules and execution errors 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 @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

CustomLabel

-0.15

scratch

-0.15

ValuePair

-0.15

jabi

-0.15

AREST

-0.14

atham

-0.14

 Legs

-0.14

Paren

-0.14

stag

-0.14

crast

-0.13

POSITIVE LOGITS

otas

0.15

 åĬł

0.14

 åĿ

0.14

azole

0.14

à¹Ĥà¸¢

0.14

azar

0.14

ÅĻej

0.14

 Modular

0.13

£

0.13

az

0.13

Activations Density 0.017%

references to programming modules and execution errors in code

No Comments

No Known Activations