INDEX

Explanations

terms related to mathematical expressions and limits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 rusty

-0.15

iless

-0.15

oref

-0.14

isle

-0.14

.addTo

-0.14

viso

-0.14

 mutually

-0.14

hek

-0.13

ombok

-0.13

 Shel

-0.13

POSITIVE LOGITS

lim

0.50

 limit

0.49

lim

0.46

limit

0.43

 Limit

0.40

Lim

0.40

 limits

0.39

Lim

0.39

.limit

0.38

 Limits

0.37

Activations Density 0.168%

terms related to mathematical expressions and limits

No Comments

No Known Activations