INDEX

Explanations

content aimed at beginner-level individuals or topics

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

êµ´

-0.15

oro

-0.15

ustomed

-0.15

 xhttp

-0.14

usi

-0.14

ephy

-0.14

ello

-0.14

 Unauthorized

-0.14

epar

-0.14

 Cres

-0.14

POSITIVE LOGITS

/basic

0.18

pong

0.17

-level

0.17

 Pole

0.15

ãĥ³ãĥĩ

0.15

-basic

0.15

.basic

0.14

PAD

0.14

 level

0.14

 Pang

0.14

Activations Density 0.048%

content aimed at beginner-level individuals or topics

No Comments

No Known Activations