INDEX

Explanations

terms related to programming concepts and document structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

_sensitive

-0.15

emen

-0.14

plor

-0.14

 Moff

-0.14

onet

-0.14

 Curtain

-0.14

ÙĦØ§ÙĦ

-0.13

 Chow

-0.13

áº«

-0.13

tej

-0.13

POSITIVE LOGITS

adu

0.21

âķĿ

0.15

ÐºÑĥÐ»

0.15

Äįas

0.15

ALSE

0.14

NE

0.14

uga

0.14

ril

0.14

ustum

0.14

Ready

0.14

Activations Density 0.007%

terms related to programming concepts and document structure

No Comments

No Known Activations