INDEX

Explanations

sequences of characters that indicate a structured format or categorization, possibly for names, addresses, or specific data points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

ninger

-0.14

-regexp

-0.14

etÃŃ

-0.13

à¸²à¸¢

-0.13

ãĥ³ãĥĩãĤ£

-0.13

lick

-0.13

 limite

-0.13

 tÃ¼rlÃ¼

-0.13

lider

-0.12

æ¨

-0.12

POSITIVE LOGITS

/linux

0.14

erez

0.14

ogan

0.13

 VÅ¡ech

0.13

eca

0.13

 mant

0.13

ision

0.12

vyÅ¡Å¡ÃŃ

0.12

ucas

0.12

emean

0.12

Activations Density 0.058%

sequences of characters that indicate a structured format or categorization, possibly for names, addresses, or specific data points

No Comments

No Known Activations