INDEX

Explanations

programming-related terms and exceptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

urile

-0.14

reu

-0.14

otros

-0.14

ÙģÙĪ

-0.13

 erotik

-0.13

 prest

-0.13

GENCY

-0.13

 porr

-0.12

Ã¡te

-0.12

 torino

-0.12

POSITIVE LOGITS

iddi

0.17

isyon

0.14

avanaugh

0.14

Ð¾Ð³Ð¾Ð´

0.13

kest

0.13

Ã¦r

0.13

aghetti

0.13

ãĥ»ãĥ»ãĥ»↵↵

0.13

Ð²ÐµÑĢÐ¶

0.13

nist

0.12

Activations Density 0.434%

programming-related terms and exceptions

No Comments

No Known Activations