INDEX

Explanations

references to online community resources and help forums

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ANGO

-0.18

-0.17

fat

-0.16

 Tales

-0.16

py

-0.15

Fat

-0.15

Py

-0.15

wg

-0.14

ango

-0.14

InputDialog

-0.14

POSITIVE LOGITS

 è¢

0.19

Trace

0.16

ified

0.16

 Overflow

0.16

Ã©ra

0.15

ibi

0.15

iable

0.15

bl

0.15

ellation

0.15

bruar

0.15

Activations Density 0.023%

references to online community resources and help forums

No Comments

No Known Activations