INDEX

Explanations

address-related information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

angelog

-0.15

 Giá»Ľi

-0.14

alim

-0.14

âĪı

-0.14

 scaleX

-0.13

ï¼ŀ

-0.13

angkan

-0.13

bsd

-0.13

 ration

-0.13

 Ð´Ð¾Ñģ

-0.13

POSITIVE LOGITS

 Suite

0.52

 suite

0.48

Suite

0.43

Ste

0.40

 suites

0.37

suite

0.37

ste

0.37

-suite

0.35

uite

0.33

STE

0.31

Activations Density 0.146%

address-related information

No Comments

No Known Activations