INDEX

Explanations

references to individual experiences and personal opinions

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

 vlastnÄĽ

-0.16

IODevice

-0.15

queda

-0.14

_UNIQUE

-0.14

Ð¾ÐºÐµ

-0.14

 Ø±Ø¶

-0.13

ourcem

-0.13

udder

-0.13

Ð¾Ð½Ð¾Ð²

-0.13

POSITIVE LOGITS

 certainly

0.92

 definitely

0.84

 sure

0.81

 Certainly

0.69

 surely

0.69

sure

0.67

 Sure

0.62

 Definitely

0.62

Sure

0.57

 definite

0.47

Activations Density 0.369%

references to individual experiences and personal opinions

No Comments

No Known Activations