INDEX

Explanations

repeated or common elements across different contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ideon

-0.16

anj

-0.15

ĥ

-0.14

aria

-0.14

issing

-0.14

ewidth

-0.13

anan

-0.13

 åĭ

-0.13

aida

-0.13

POSITIVE LOGITS

/Resources

0.14

StackSize

0.14

akte

0.14

Steam

0.14

urb

0.14

 Milli

0.14

_disk

0.14

intro

0.14

MAP

0.14

 :č↵

0.13

Activations Density 0.013%

repeated or common elements across different contexts

No Comments

No Known Activations