INDEX

Explanations

references to the timing and context of events

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

 Reform

-0.15

 credited

-0.14

itten

-0.13

 Reed

-0.13

enth

-0.13

Ã¶h

-0.13

readcr

-0.13

du

-0.13

Bas

-0.13

 alter

-0.12

POSITIVE LOGITS

 writing

0.40

writing

0.31

 Writing

0.28

Writing

0.27

 publishing

0.27

-writing

0.26

 publication

0.26

 write

0.24

 press

0.24

 typing

0.24

Activations Density 0.036%

references to the timing and context of events

No Comments

No Known Activations