INDEX

Explanations

numerical data and symbols commonly associated with programming or technical content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

åį

-0.18

 Challenge

-0.14

 challenge

-0.14

'gc

-0.14

Challenge

-0.14

 challenges

-0.14

posium

-0.14

unc

-0.14

sol

-0.14

NAS

-0.14

POSITIVE LOGITS

isch

0.16

ernity

0.15

ibia

0.15

arov

0.14

rors

0.14

akh

0.14

USR

0.14

edor

0.13

Kem

0.13

Multiplicity

0.13

Activations Density 0.006%

numerical data and symbols commonly associated with programming or technical content

No Comments

No Known Activations