INDEX

Explanations

punctuation and formatting elements in the text

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

-0.17

oyo

-0.15

 æį

-0.14

elf

-0.14

Kul

-0.14

islav

-0.14

 screw

-0.14

Į¨

-0.14

Api

-0.14

 track

-0.14

POSITIVE LOGITS

anst

0.17

uve

0.16

 Ø§ÙĦØ±ÙħØ²ÙĬØ©

0.16

 ifndef

0.15

SAX

0.15

acher

0.15

ÑĢÑıÐ´

0.15

çĻ

0.15

baugh

0.15

agner

0.15

Activations Density 0.068%

punctuation and formatting elements in the text

No Comments

No Known Activations