INDEX

Explanations

instances and phrases that indicate a beginning or introduction to a topic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

¢åįķ

-0.15

çļĦæĺ¯

-0.15

avin

-0.14

ipp

-0.14

idl

-0.14

 spite

-0.14

 Antar

-0.14

verter

-0.13

wei

-0.13

Î¶Îµ

-0.13

POSITIVE LOGITS

ologne

0.17

sum

0.17

 thus

0.16

ej

0.15

cade

0.15

ubble

0.14

Sen

0.14

thus

0.14

leton

0.14

Activations Density 0.093%

instances and phrases that indicate a beginning or introduction to a topic

No Comments

No Known Activations