INDEX

Explanations

instances of the word "By," typically indicating authorship or attributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

upal

-0.15

Ð¸Ð¼Ð²

-0.14

enaire

-0.14

ãĤ¤ãĥ³ãĥĪ

-0.14

itech

-0.14

 Morrow

-0.14

ÙĪÙħØ§Øª

-0.14

duino

-0.14

emble

-0.14

ima

-0.13

POSITIVE LOGITS

0.19

 scre

0.15

zig

0.14

ickers

0.14

edm

0.14

0.13

ogg

0.13

 turned

0.13

Nic

0.13

Activations Density 0.009%

instances of the word "By," typically indicating authorship or attributions

No Comments

No Known Activations