INDEX

Explanations

instances of the word "created" and variations of it, indicating an emphasis on origin or authorship

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

 transl

-0.15

/mac

-0.15

 tack

-0.15

aucoup

-0.15

ych

-0.15

 Bent

-0.14

urr

-0.14

 Creative

-0.14

stick

-0.14

POSITIVE LOGITS

andest

0.18

"default

0.17

avia

0.16

ibus

0.15

Charset

0.15

berger

0.14

bem

0.14

morph

0.14

à¥Įà¤²

0.14

Activations Density 0.031%

instances of the word "created" and variations of it, indicating an emphasis on origin or authorship

No Comments

No Known Activations