INDEX

Explanations

specific instances of the word "the" along with mentions of quantifiable data related to various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ernal

-0.16

/current

-0.15

 Lens

-0.15

est

-0.15

emon

-0.14

iston

-0.13

ãģ£ãģ

-0.13

 Leader

-0.13

_EOF

-0.13

deme

-0.13

POSITIVE LOGITS

mmo

0.17

oplevel

0.16

 effort

0.16

enment

0.15

 âĹĦ

0.15

usi

0.14

opers

0.14

inis

0.14

æľĹ

0.14

 spoj

0.13

Activations Density 0.038%

specific instances of the word "the" along with mentions of quantifiable data related to various subjects

No Comments

No Known Activations