INDEX

Explanations

programming concepts related to classes and abstract data types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

aidu

-0.15

 Vand

-0.14

erli

-0.14

 paran

-0.14

-mf

-0.14

olian

-0.14

edBy

-0.14

idth

-0.13

Dump

-0.13

 Dump

-0.13

POSITIVE LOGITS

é¨

0.17

aleza

0.16

ignon

0.16

 bitmask

0.15

zy

0.15

 Incre

0.15

rag

0.14

 Ø§ÙĦÙĨÙĩ

0.14

anine

0.14

baum

0.13

Activations Density 0.005%

programming concepts related to classes and abstract data types

No Comments

No Known Activations