INDEX

Explanations

word patterns or repetitions, especially involving the letters "by" and suffixes like "ce."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

eln

-0.15

æ¢¨

-0.15

ackbar

-0.15

ÏĦÎ¿Ïħ

-0.14

baugh

-0.14

agal

-0.14

tero

-0.14

 Rowe

-0.14

ecom

-0.14

ula

-0.14

POSITIVE LOGITS

 Laud

0.16

à¥įà¤µà¤µ

0.15

inbox

0.14

 PROT

0.14

ieu

0.14

elah

0.14

Ung

0.14

nia

0.14

Dt

0.14

 Pest

0.14

Activations Density 0.005%

word patterns or repetitions, especially involving the letters "by" and suffixes like "ce."

No Comments

No Known Activations