INDEX

Explanations

references to annotations or metadata in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lopedia

-0.18

vie

-0.16

cular

-0.15

rana

-0.15

 secular

-0.15

LLU

-0.14

/***/

-0.14

 massaggi

-0.14

elize

-0.14

NotFoundError

-0.14

POSITIVE LOGITS

param

0.24

var

0.24

see

0.21

 param

0.20

var

0.18

throws

0.18

api

0.18

see

0.18

return

0.18

author

0.16

Activations Density 0.003%

references to annotations or metadata in a document

No Comments

No Known Activations