INDEX

Explanations

ellipsis or trailing punctuation indicating pauses or omissions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

 paternal

-0.20

.AddParameter

-0.15

itus

-0.14

ç³»

-0.14

aday

-0.14

 undermin

-0.14

 Peaks

-0.14

çľĭçĿĢ

-0.13

çľ¼

-0.13

Wir

-0.13

POSITIVE LOGITS

 Morav

0.17

lingen

0.16

emean

0.16

á»ı

0.16

çĵ¦

0.15

ondheim

0.15

ierge

0.15

astro

0.15

ntax

0.14

Ø²Ø§ÙĨ

0.14

Activations Density 0.029%

ellipsis or trailing punctuation indicating pauses or omissions

No Comments

No Known Activations