INDEX

Explanations

instances of numerical values and associated metrics

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

er

-0.20

que

-0.18

ines

-0.17

ed

-0.17

ly

-0.16

oke

-0.15

 Kaplan

-0.15

in

-0.15

POSITIVE LOGITS

JKLM

0.15

sdk

0.14

-thumbnails

0.14

tÃ¼

0.14

#=

0.14

Globals

0.14

_nat

0.14

atÃ¼rk

0.13

ÑĨÐ²ÐµÑĤ

0.13

MediaType

0.13

Activations Density 0.022%

instances of numerical values and associated metrics

No Comments

No Known Activations