INDEX

Explanations

numerical and statistical data

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

Negative Logits

ackle

-0.16

ODE

-0.15

agues

-0.15

elah

-0.14

/Area

-0.14

issement

-0.14

HLT

-0.14

obil

-0.14

stants

-0.13

à¹ģà¸Ī

-0.13

POSITIVE LOGITS

ToArray

0.15

ToObject

0.14

OrElse

0.14

iou

0.14

Filled

0.14

 pred

0.13

WithTag

0.13

_duplicates

0.13

 filled

0.13

 tempo

0.13

Activations Density 0.050%

numerical and statistical data

No Known Activations